Sei sulla pagina 1di 213

ECOSYSTEM SERVICES IN PLANNING PRACTICE

FOR URBAN AND TECHNOLOGICALLY ADVANCED LANDSCAPES

BY

JORDI C. HONEY-ROSS

DISSERTATION

Submitted in partial fulfillment of the requirements


for the degree of Doctor of Philosophy in Regional Planning
in the Graduate College of the
University of Illinois at Urbana-Champaign, 2012

Urbana, Illinois

Doctoral Committee:

Professor Daniel W. Schneider, Chair


Associate Professor Nicholas Brozovic
Professor Edward Feser
Professor Pedro Arrojo Agudo, University of Zaragoza, Spain
ABSTRACT
Research on ecosystem services strives to build stronger linkages between ecological and

economic systems in order to improve ecosystem management and human well-being. By clarifying how

human populations benefit from ecosystems we may protect valuable ecosystem functions and improve

human welfare more strategically. To advance the field of ecosystem services my dissertation asks three

questions:

(1) How can historical research inform why ecosystem based management approaches have been

integrated or ignored into watershed management?

(2) What is the relationship between technological innovation and the value of ecosystem services?

(3) Can the restoration of ecosystem structures and functions influencing stream temperature provide

valuable ecosystem services for water treatment managers?

I focus on ecosystem services in the Llobregat watershed near Barcelona, Spain, where over 3

million residents rely on the Llobregat River for basic water needs. The withdrawal of drinking water

from the Llobregat River has created clear linkages between local well-being and the rivers aquatic

ecosystem. Two water treatment facilities withdraw water from the Llobregat, and both have recently

installed desalination technology. However the new treatment systems are expensive to operate. In this

urban and technologically advanced context, I explore how ecosystem services may be managed to meet

environmental and economic goals.

Chapter 1 reviews the literature on ecosystem services, describes the study area, and begins to

outline my argument developed in the dissertation. The number of articles published on the topic of

ecosystem services has exploded in the last decade, and in this chapter I describe how my research

contributes to this discussion. One major argument pertains to the sequence in which the field usually

studies the linkages between ecological and economic systems. Most research begins with the

biophysical; and later draws on the social sciences to attach monetary values to ecosystem structures or
ii
functions. Instead, I propose that we study the economic system first, including its decision-making and

technological context, and then allow our ecological research to follow from that insight.

Chapter 2 consists of an environmental history that examines the individuals and institutions that

have controlled the flow of water in the Llobregat River, and by extension, dictated water management

practices. The study of ecosystem services is often presented as a new and innovative framework for

addressing environmental problems. Through archival research and personal interviews, I examine how

historic resource users have integrated or ignored this framework for decision making in the Llobregat

watershed. I find that throughout the twentieth century, most river managers did not adopt an ecosystem

services approach, but rather favored hardscaped and structural solutions. At the same time I find that

proposals to manage ecosystem services related to flood control were proposed as far back as 1890, and

these ideas were implemented in the 1930s. Notions of ecosystem services have more than a century of

history in the watershed, and yet they have not played a central role in river and water management.

Structural approaches have dominated decision making and continue to play a dominant role, as most

recently demonstrated by the installation of desalination treatment plants. The recommendations from the

Commission for the Study of Salinity of the Llobregat River (CESALL) in 1932 outlined structural

solutions for addressing the Llobregat Rivers water pollution and supply problems, and these

recommendations had a lasting impact for the remainder of the twentieth century. I argue that our current

system of water treatment and distribution reflects a historical tradition that has favored technological

fixes when addressing water management problems.

Chapter 3 examines the value of urban ecosystem services when sophisticated technology

mediates our relationship with ecosystem processes. Most research on ecosystem services is being

conducted in natural or pristine areas, while less attention has been directed at urban ecosystem services

and their relationship with technological change. Understanding the relationship between technological

change and the value of ecosystem services is relevant because it may broaden the circumstances in which

ideas about ecosystem services may be implemented in practice. I argue that the expected tradeoff
iii.
between natural and manufactured capital is false. Rather, the adoption of new technologies is

complementary to ecosystem management. This point is illustrated with a case study that analyzes how

the installation of sophisticated drinking water treatment technology increased the value of ecosystem

services in Barcelona, Spain. The implication is that the supply of ecosystem services is not fixed; and

technological change will reshape which ecosystem services are valuable but not obviate the need for

them entirely.

Finally, in Chapter 4 I use a biophysical stream temperature model to assess how increases in

riparian vegetation and stream discharge may reduce stream temperatures and treatment costs

downstream. This chapter serves as an example of the type of research that I am proposing for the field.

The type of ecological modeling I chose emerged from an analysis of the technological and decision-

making context of ecosystem users. By modeling different ecological scenarios, I find that existing

forests along the Llobregat River save water treatment managers 79,000 per year, while the restoration

of additional riparian forests could generate economic savings in the range of 57,000- 156,000 per year.

Stream restoration at higher elevations would yield greater benefits than restoration in the lower reaches.

Moderate increases in stream discharge (25%) could generate savings of 40,000 per year. This research

underscores that ideas about ecosystem services have the potential to be more widely adopted if research

focuses on the demand for these services rather than the supply.

iv.
Acknowledgements

This dissertation would not have been possible without the support from my family, friends and

colleagues. I thank my advisor, Daniel Schneider, for allowing me to pursue my own research path while

also keeping me on track when I went astray. Nick Brozovic provided insightful commentary every step

of the way and displayed extraordinary patience when I came with questions. Similarly, I benefited from

several discussions with Edward Feser and Pedro Arrojo, both of whom made constructive and perceptive

comments during the defense.

Lew Hopkins and Elizabeth L. Sweet made it possible for me to join the doctoral program at the

Department of Urban and Regional Planning at the University of Illinois. The University awarded me a

fellowship during my first year, and later offered me a Teaching Assistantship, a Dissertation Travel

Grant, a Tinker pre-doctoral research grant, and a fellowship from the Human Dimensions of

Environment Systems program. I also was fortunate to be granted a Research Assistantship under the

direction of Professor Kathy Baylis.

Early on, when searching for a dissertation topic, I relied on guidance from experts in the field of

urban planning and water management. I was fortunate to be granted interviews from Narcs Prat, Enric

Tello, Germ Bel, Rafael Mujeriego, Oriol Nel.lo, David Saur, Josep Anton Acebillo, Marc Monlle,

Ignasi Puig, and Jaume Freire.

When my attention turned toward the Llobregat River, those familiar with the waterway

welcomed me without hesitation. These stewards shared with me their ideas and aspirations, all of which,

in some way, have helped me develop a stronger research project. In particular I would like to thank

Roger Lloret, whom I met while preparing for a backpacking trek down the river in July 2008. Roger

taught me about water treatment, water chemistry, and the rivers history. He also arranged important

interviews with numerous co-workers and friends. I also thank ngel Miralda for accompanying me

during my hike down the river for two days, as well as Hctor Oliva and Sara Ortiz Escalante for also

v.
joining me on the adventure. Ricard Sosa, Llus Canals and Rafael Fernndez from the Catalan Water

Agency (ACA) gave me a fantastic tour of the Baells Dam, even though my visit was unannounced. Neus

Santamaria introduced me to the magical history of the company towns along the Llobregat River. Josep

Ribera, David Hernndez, Eloi Escud, Marc Vinyals and the committed community organizers at Prou

Sal! brought me up to speed with the river pollution controversy in Sallent. Their work largely inspired

me to write about the salinity conflict in the Llobregat River. In Martorell, Rafa Diez and Roger

Arqu from Martorell Viu accompanied me along the Anoia stream and its confluence with the Llobregat.

Joan Bordas and Pep Clavero kindly spent a morning teaching me about the plant life in the lower

Llobregat River. Mia Morante, Pau Fortuo and Nuria Snchez gave me a lesson in macro-invertebrates

and biological indicators as we collected water quality samples along the Llobregat and its tributaries.

And Josep Ferret generously granted me an interview, shared articles about the Llobregat River written by

his grandfather in the 1930s, and gave me two books on the history of the river, its aquifer and floods.

Major funding to complete this research project was generously provided by Centre Tecnolgic

de lAigua (CETaqua) and the Catalan Water Agency (ACA), with assistance from the Catalan Institute

for Water Research (ICRA). This grant would not have been possible without the unwavering support

from Carlos Campos, Joana Tobella, and Montserrat Termes at CETaqua; and Antoni Munn and Gabriel

Borrs at ACA. I am deeply grateful for the confidence they deposited in me to execute a project which,

at the time, was merely an idea. Similarly, Ignacio Escudero and Josep Llus Armenter at Aiges de

Barcelona (AGBAR) allowed me to consult sensitive water treatment data.

At the AGBAR water treatment facility in Sant Joan Desp, I received valuable assistance from

lex Vega, who helped me find my way within the water treatment facility. I also thank Marc Pons,

Albert Teuler, Antoni Bernal, Christian Sols, Josep Planas and the entire team at the AGBAR-Sant Joan

Desp laboratory.

At CETaqua I would like to thank Benoit Lefvre, Jose Luis Cortina, Oriol Gibert, Susana

Gonzlez, Xavier Bernat, Catalina Balseiro, Rosa Maria Pieras, Isabel Escaler, and Carlos Montero.
vi.
At ATLL I thank Fernando Valero, ngel Barcel and Ramon Arbs for their patience when

explaining to me the complexities of the EDR desalination system. Under exceedingly difficult

circumstances, they have shown to be model public servants.

Many at ACA have shared valuable data or insight with me. Most notably, I would like to thank

Mnica Bardina, Llus God, Jordi Rovira, Josep Maria Nierola, Carles Cardona, Rosa Serrano, Carlos

Barbero, Jordi Molist, Joan Verd, Victoria Colomer, and Evelyn Garcia.

The research team at the Catalan Institute for Water Research (ICRA) generously gave me

institutional support during my time in Catalonia. Sergi Sabater and Vicen Acua were early believers

in the project. Throughout late 2009 and early 2010, when the institutional support was stuck in an

impasse, they displayed extraordinary perseverance and reasserted their support for this project. Once I

was invited to join ICRA, I benefited from discussions with Rafael Marc, Marta Terrado, Roberto

Merciai, Daniel von Schiller, Elisabet Torns, Marta Ricart, Ldia Ponsat, Xisca Timoner and Ramon

Batalla. Carmen Gutierrez, Olga Corral, Emma Collelldevall, Anna Cornella, Jaume Alemany, Ivn

Snchez and Dami Barcel also provided essential support at ICRA.

Emi Turull at the Museu de les Aiges generously gave me several books on the history of

AGBAR and water management in Barcelona. I am grateful that this productive exchange with the

Museum has continued with Sonia Hernndez and Rosa Prat. I had illuminating conversations about the

historical research with Santi Gorostiza and Albert Fbrega Enfedaque, both of whom generously shared

with me many of their findings. Esteve Torrens Prez de los Cabos provided me with the detailed

description of his grandmother, Antnia Burs Borrs, in addition to family photos.

Jordi Badia, Florenci Valls and Sergi Falguera taught me about the geology of the Llobregat

watershed. Albert Soler gave me a fantastic field visit of the Cardener River, the restoration work at

Vilafruns, and an explanation of groundwater flows. Jaume Ribera also took me on an informative hike

along the Cardener River in Sria, and Josep Illa did not hesitate to provide me with background

information about the restoration of riparian forests. Antonio Palacios from the rea Metropolitana de
vii.
Barcelona provided valuable comments on work in progress. Annelies Brockman, Quim Prez, Daniel

Barb, and Jaume Descls oriented me with regard to the social issues and political tug of wars over

water management in Catalonia. lvaro Martnez-Novillo Gonzlez paid for my computer repair after an

opening car door knocked me off my bicycle.

I would also like to thank those who inspired me to pursue my doctoral studies in the first place,

especially Lincoln P. Brower, Isabel Ramrez, Carlos Galindo-Leal, Lee Pagni, William Toone, Roger

Paredes, Judith Serna and Daniel Cams.

Lastly, I thank my entire family for their unconditional support.

viii.
Table of Contents
CHAPTER 1. INTRODUCTION ................................................................................................... 1
1.1 State of the Literature on Ecosystem Services .................................................................................... 1
1.2 Study Area ........................................................................................................................................ 10
CHAPTER 2. ENVIRONMENTAL HISTORY OF THE LLOBREGAT WATERSHED ............................ 17
2.1 Introduction ....................................................................................................................................... 19
2.2 Desalination of the Llobregat River .................................................................................................. 22
2.3 Salinity Conflict ................................................................................................................................ 26
2.4 Ecosystem Services for Flood Management ..................................................................................... 66
2.5 Conclusion ........................................................................................................................................ 75
2.6 Acknowledgements ........................................................................................................................... 80
2.7 References Chapter 2 ........................................................................................................................ 81
CHAPTER 3. URBAN ECOSYSTEM SERVICES AND TECHNOLOGICAL CHANGE .......................... 83
3.1 Introduction ....................................................................................................................................... 84
3.2 Urban Ecosystem Services ................................................................................................................ 85
3.3 Ecosystem Services and Technology ................................................................................................ 88
3.4 Case Study: Water Treatment Technologies in Barcelona, Spain .................................................... 94
3.4.1 Advanced Membrane Technologies for Water Treatment ......................................................... 95
3.4.2 The Value of Ecosystem Services Following Technological Change ....................................... 99
3.5 Discussion ....................................................................................................................................... 107
3.6 Acknowledgements ......................................................................................................................... 110
3.7 References Chapter 3 ...................................................................................................................... 111
CHAPTER 4. MANAGING ECOSYSTEM SERVICES TO MEET S TREAM TEMPERATURE OBJECTIVES
IN THE LLOBREGAT R IVER , S PAIN ...................................................................................... 115
4.1 Introduction ..................................................................................................................................... 116
4.2 Study Area ...................................................................................................................................... 119
4.3 Methods........................................................................................................................................... 120
4.4 Results ............................................................................................................................................. 125
4.5 Discussion ....................................................................................................................................... 140
4.6 Acknowledgements ......................................................................................................................... 144
4.7 References Chapter 4 ...................................................................................................................... 145

ix.
APPENDICES ...................................................................................................................... 148
Appendix A. Osmotic Pressure, Total Dissolved Solids, & Energy Efficiency at AGBAR................. 149
Appendix B. Electrodialysis Reversal (EDR), Conductivity, & Energy Efficiency at ATLL .............. 153
Appendix C. Heat Fluxes Modeled in the Stream Network Temperature Model (SNTEMP) ............. 154
Appendix D. Stream Network Model Execution .................................................................................. 155
Appendix E. Input Files for the Stream Network Model ...................................................................... 157
Appendix F. Field Work for Shading Estimates ................................................................................... 165
Appendix G. Calibration of SNTEMP Model ...................................................................................... 166
Appendix H. Full Results for SNTEMP ............................................................................................... 167
Appendix I. Notes on Scale Selection for SNTEMP ............................................................................ 172
Appendix J. Stream Segment Model (SSTEMP) .................................................................................. 175
Appendix K. Stream Network Model from Castellbell to Abrera ........................................................ 183
Appendix L. Riparian Vegetation and Restoration Costs ..................................................................... 189
REFERENCES ...................................................................................................................... 192

x.
CHAPTER 1. INTRODUCTION
1.1 State of the Literature on Ecosystem Services
People and cities depend on the goods and services produced by our planets ecosystems. This

dependent relationship between human well-being and the biophysical world is encapsulated by the

relatively new notion of ecosystem services. The food we eat, the air we breathe, and the water we drink

all derive from ecosystem processes. However our dependence on these ecosystems has not prevented us

from stressing them and reducing their capacity to meet our needs (MA 2005). To maintain our valuable

ecosystem services intact, we must improve resource management and decision making. The ecosystem

services framework promises to bring together the ecological and social sciences to meet this challenge.

Journals are publishing new ideas about ecosystem services at an extraordinary rate. Within the

last six years, five leading publications have released special issues on ecosystem services: Ecology and

Society (Volume 11, No 2, 2006); Ecological Economics (Volume 65, Issue 4, May 2008); the

Proceedings of the National Academy of Sciences (Volume 105, No 28, July 2008); Frontiers in Ecology

and the Environment (Volume 7, Issue 1, February 2009); Ecological Economics (Volume 69, Issue 6,

April 2010); Ecological Complexity (Volume, 7 Issue 3, September 2010); and again Ecological

Economics (Volume 69, Issue 10, September 2010). In 2012, a new journal will be released with the title

Ecosystem Services. Furthermore, the National Research Council issued a comprehensive report on

Ecosystem Services entitled Valuing Ecosystem Services: Toward Better Environmental Decision Making

that has synthesized this emerging discussion (NRC 2005). And internationally, the United Nations

Millennium Ecosystem Assessment adopted the ecosystem services framework to assess the state of the

globes ecosystems (MA 2005).

The most well-known categorization of ecosystem services divides them into four groups: (1)

provisioning services, which refers to ecosystems ability to generate essential goods such as timber, fuel,

food and fiber; (2) regulating services which refers to the regulation of climate or the water cycle; (3)

supporting services, such as pollination, population control, soil formation, and other ecological
1.
properties upon which life depends; and finally (4) cultural services, which provide humans with

recreational, spiritual and aesthetic values (MA 2005). This categorization of ecosystem services, while

useful conceptually, has been difficult to operationalize. Problems arise when these four services are

valued as separate entities because in practice, several services may overlap and therefore lead to double

counting (Boyd and Banzhaf 2007). To address this problem, ecosystem services have been re-defined as

the components of nature, directly enjoyed, consumed, or used to yield human well-being (Boyd and

Banzhaf 2007). This definition makes a distinction between intermediate components and final services

and also makes it easier to integrate natural capital into a system of national accounts. Building on this

notion, Fisher (et al. 2009) further simplified the definition of ecosystem services as the aspects of

ecosystems utilized (actively or passively) to produce human well-being.

As these definitions suggest, the field of ecosystem services strives to build stronger linkages

between ecological and economic systems with the purpose of improving ecosystem management and

human well-being (Daily 1997, Arrow et al. 200, Farber et al. 2006). Research on ecosystem services is

quintessentially interdisciplinary because it weaves together the physical, biological, and social sciences

into a framework for analysis. The successful integration of these fields into a coherent, concrete and

practical management approach has the potential to transform environmental policy at several scales.

Proponents of ecosystem services have argued that this framework offers the most promising way forward

for the field of conservation biology (Chan et al. 2006, Armsworth et al. 2007). Protecting our life support

systems has also resonated with advocates for the global poor (Martnez-Alier 2002), because populations

in less developed countries are more dependent on ecosystem services and more vulnerable to their loss

(Sachs and Reid 2006). Major research institutions, conservation organizations, foundations and the

private sector are investing in research on ecosystem services (Montenegro 2008, Stanton et al. 2010,

Armsworth et al. 2011). These groups expect that a breakthrough in the field will open up new avenues

for solving sustainability challenges.

2.
The concept of ecosystem services was developed by conservation biologists (Ehrlich and

Mooney 1983, Daily 1997) in collaboration with environmental economists (Costanza et al. 1997) to find

new ways to protect biodiversity. The legacy from biodiversity conservation has left an imprint on the

field, as researchers are far more likely to advocate for an ecosystem services approach in the biologically

rich, yet economically poor countries in the developing world rather than in industrialized nations

(Norgaard 2010). Research on ecosystem services took off in part because it held the potential to tie

together development and conservation objectives.

At the same time, it has been acknowledged that ecosystem services are merely a new way of

talking about an old idea. Economists have studied ecosystem services as positive and negative

externalities, and it is well known that ecosystems generate valuable goods and services that are not

accounted for in the market (Pigou 1920, Krutilla 1967, Costanza and Daly 1992). However advocates for

research in ecosystem services suggest that the conceptual frameworks used in the past are insufficient to

manage ecosystem services today (Daily and Matson 2008, Liu et al. 2010). Furthermore, the ecological

processes that underpin ecosystem services remain poorly understood (Kremen and Ostfield 2005,

Kremen 2005). For example, it is well know that mangroves provide habitat for shrimp production

(Aburto-Oropeza et al. 2008), or that rivers purify waters (Grimm et al. 2005), and yet it remains

exceedingly difficult to calculate the marginal values associated with restoring mangrove habitat or

restoring river ecosystems because we have not successfully unified the language of ecology with the

social sciences (de Groot et al. 2002, Fisher et al. 2008).

Conservation organizations were the first to use the idea of ecosystem services to advance their

goals of habitat conservation (Pagiola, Bishop, and Landell-Mills 2002, Tallis et al. 2009). In particular,

forest conservationists first began to exploit the funding opportunities that ecosystem services could bring

to forest dwelling communities (Bishop and Landell-Mills 2002). In the late 1990s and early 2000s, the

idea of payments for ecosystem services attracted attention from government officials and conservation

donors working in the developing world. Costa Rica became the first nation to pay land owners for forest
3.
conservation based on the idea that their forests provided valuable services to the country at large (Pagiola

2008). Mexico and other countries soon followed suit with their own payment for environmental services

program, often focusing on the protection of water supplies (Muoz-Pia et al. 2008, Stanton et al. 2010).

Payments for Ecosystem Services (PES) are defined as voluntary transactions where a well defined

ecosystem service is bought by a buyer if the provider secures a provision of a service (Wunder 2005).

Thus ecosystem services provided an economic rationale to invest in conservation, while at the same time

it aligned development and conservation goals. Furthermore, the ecosystem services approach, one that

bound together ecosystem beneficiaries with ecosystem service providers in a mutually beneficial

financial transaction, promised to open up new opportunities for conservation. Paying for conservation

radically differed from the existing conservation methods used to date, and quickly became seen as a

viable alternative to the legal strategies, or development projects, that had been the standard conservation

tools. Moreover, new arguments surfaced that direct economic payments would be economically more

efficient than alternative conservation schemes (Ferraro and Kiss 2002, Ferraro and Simpson 2002),

generating even more discussion about the need to quantify and understand ecosystem services.

While payments that pay for ecosystem services have been implemented in the developing world,

more prosperous nations rarely rely on the idea of ecosystem services to manage their ecosystems. With

the exception of conservation payments for agriculture (Dobbs and Pretty 2008), or the program on

compensatory wetland mitigation (BenDor and Brozovic 2007, Palmer and Filoso 2009, BenDor and

Doyle 2010), examples in the urban and developed world remained scant. The most prominent

application was New York Citys drinking water protection program in the Catskill watershed. Yet when

policy makers tried to replicate New Yorks experience, it was found that water managers in most major

cities of the United States had already installed the filtration technology that New York City successfully

avoided. This outcome only reaffirmed the idea that ecosystem services came from predominantly

untouched areas, such as the Catskill watershed, and managers would have fewer opportunities to adopt

an ecosystem approach once a new technology was put in place.


4.
Also in the late 2000s, it became increasingly evident that the excitement surrounding ecosystem

services has not been accompanied by a parallel transformation in how ecosystems were managed. Even

prominent scholars recognized that it was time to deliver (Daily et al. 2009). Critics raised questions

about why more success stories have not been found (McCauley 2006). To encourage a speedier adoption

of the ecosystem services framework in practice, proponents began to advocate for research that could

help decision-makers implement these ideas with resource managers (Cowling et al., 2008; Daily and

Matson, 2008; Daily et al., 2009; Muradian et al., 2010).

Throughout the dissertation, I argue that two major factors have limited the application of ideas

about ecosystem services. First, researchers have ignored the opportunities to find and exploit ecosystem

services in urban and technologically advanced landscapes (Chapter 3). Second, research has not

dedicated sufficient attention to studying the demand for ecosystem services (Chapter 4). Instead, most

research has focused mostly on the biophysical processes, while leaving the economic analysis and

decision-making context as secondary. This dissertation addresses these limitations by studying the

demand for ecosystem services in an urban and technologically sophisticated environment.

Of course, this is not the only gap in the literature. There are other obstacles that also must be

addressed to move ideas about ecosystem services from theory to practice. According to the special

report on ecosystem services released by the National Research Council, one fundamental challenge in

the field of ecosystem services lies in linking ecosystem structure and function with economic values

(NRC 2005). Ecosystem structure refers to the composition of the ecosystem and the physical and

biological organization defining how these parts are organized, while ecosystem function describes a

process that takes place in an ecosystem as a result of interactions of plants, animals and other

organisms (NRC 2005). Both ecosystem function and structure are value free descriptors that provide

ecosystem services when they generate human well-being. Whether or not ecosystem function and

structure are ecosystem services often depends on seasonal fluctuations or where the ecosystem is located

relative to human populations. For example, the capacity of wetlands to treat sewage, mitigate floods, or
5.
purify water, often depends on their proximity to human populations as well as seasonal fluctuations

(Brauman et al. 2007).

The subtle, yet critical differences between ecosystem functions/structure; ecosystem service, and

economic value have been the source of confusion. Their similarity may also have contributed to the

simplifying assumption that these three processes are linearly related. For example, it has been assumed

that growing mangrove habitat (ecosystem structure), will increase the populations of edible fish

(ecosystem service) that can be harvested, sold and consumed (economic value). Assuming a linear

relationship between ecosystem functions and value or between ecosystem condition and service can lead

to misleading results. Note that the ecological production functions link ecosystem structure/function

with ecosystem services, while ecosystem services are linked to economic values through economic

valuation functions (Fig 1.1).

Figure 1.1 Ecosystem function and structure have nonlinear relationships with the ecosystem services
they provide, and this relationship is described by ecological production functions. Similarly, the values
of these ecosystem services also fluctuate, and this relationship is described by economic valuation
functions.

Furthermore, spatial and temporal variability add two additional levels of complexity (Fig 1.2).

The values of any ecosystem service can change depending on its location or season. The combination of

these complexities has prevented us from acting on ideas about ecosystem services. In practical terms,

this means that resource managers do not have the means to evaluate the impacts of their decisions on the

ecosystem services in question.

6.
Figure 1.2 The chain linking ecosystem functions/structure, with ecosystem services and economic value
increases in complexity when we consider spatial and temporal variability.

Similarly, Koch et al. (2009) studied how coastal habitats provided the ecosystem service of

coastal protection. They demonstrated how valuation results varied significantly according to spatial and

temporal fluctuations such as tidal changes, season, latitude, and spatial distribution. Also, Aburto-

Oropeza et al. (2008) used a regression model to study the relationship between catch yields and the

presence of mangrove habitat in the Gulf of California in a large data set between 2001 and 2005. They

found that fish catch increased by the square root of the area of mangrove habitat. All three studies

suggest that by incorporating nonlinear relationships between ecosystem functions and values we can

offer more precise estimates as to the true utility of ecosystem services.

Social scientists have also turned their attention to non-linearities and thresholds. Repetto (2006)

suggests that US environmental policy advanced primarily during short periods of rapid change. This has

led political scientists to inquire into the conditions that may allow rapid policy change and

implementation. Similar to their ecological counterparts, they see the value in understanding the location

of thresholds and their triggers.

7.
Before we can expect decision makers to act upon our understanding of complex ecological and

economic systems, first we must identify and measure the flows of ecosystem services. This process of

ecosystem service identification leads to a significant way in which this research distinguishes itself from

existing studies on the subject. My point of departure is the demand for ecosystem services instead of the

supply. Most studies on ecosystem services begin with a particular ecosystem of interest: the coastal

mountains of Northern California (Chan et al. 2006), the montane forests on Hawaiis Kona islands

(Goldstein 2007), the coastal habitat of Thailand (Barbier et al. 2008) or the mangrove forests of Baja

California (Aburto-Oropeza et al. 2008). Having selected an ecosystem of inquiry, researchers proceed to

describe the services generated by that ecosystem. In most cases, the local populations are passive or

unaware recipients of the services provided by the ecosystem of interest, and new values are attributed to

lands that previously were not valued in the market. The unspoken assumption behind these studies is that

if the local populations were aware of the services, they would seek to protect them.

Instead of following the mainstream approach, in which one first characterizes the range of

ecosystem services in the study area, I start by studying the economic and technological system that

generates the demand for a particular set of ecosystem service. In other words, rather than beginning with

the supply of existing services, I begin with the demand, or the users objectives. I start with the context

in which decisions are made and then identify ecosystem functions that can help resource users meet their

management goals. This allows me to target my ecological research to the specific needs of ecosystem

users. Identifying the demand for ecosystem services should also facilitate the adoption of these ideas in

practice.

The inability to act on ideas about ecosystem services is a major criticism of the field. But the

stuy of ecosystem services has been criticized for other reasons as well. To begin with, discussions about

ecosystem services often latch on to romantic notions of pristine areas or wilderness, which have

been discredited by environmental historians (Cronon 1995), geographers (Castree 2005) and ecological

scientists (Kareiva et al. 2007). And when programs pay landowners for services provided by ecosystems,
8.
the distributional outcomes are not exactly what one might anticipate, as marginalized communities may

disproportionately be excluded from conservation benefits (Corbera et al. 2007). Furthermore, research

on ecosystem services quantifies and commodifies our world (Kosoy and Corbera 2010). An obsession

with the utilitarian contribution of ecosystems to human well-being, especially with quanitifcation and

commodification, might lead us to ignore the larger complexities involved with functioning ecosystems

(Norgaard 2010). Norgaard has thus criticized the notion of ecosystem services a complexity blinder.

This criticism fits into a larger critique of centralized decision-making, objective measurement and the

hubris of rational planning and reductionism (Scott 1998).

Others criticize the field of ecosystem services for placing undue emphasis on the economic

arguments for conservation, and in this way, pushing aside the ethical or rights-based arguments to

motivate environmental protection (MCauley 2006, Child 2007). After all, human self-interest and market

forces are precisely what have caused ecosystems to be modified to begin with. These critics are wary that

the economic arguments may be less appealing when the calculations point in the other direction.

According to this view, what society needs is a stronger conservation ethic and less commodification, not

more (McCauley 2006). Certainly, even the most ardent defenders of ecosystem services must recognize

that the field of ecosystem services is anthropocentric, instrumental and somewhat reductionist. And yet

the ethical motivations for conservation have never been sufficient to change human behavior in the past,

and it is unlikely that this will change any time soon. At least this new integration with economics can

provide additional motivation for improving ecosystem management (Reid et al. 2006, Costanza 2006).

Despite these criticisms, the field of ecosystem services is growing (Fisher et al. 2009). The most

recent research analyzes tradeoffs between services (Kareiva et al. 2007; Raudsepp-Hearne, Peterson, and

Bennett 2010; Chan, Hoshizaki and Klinkenberg 2011). And advances in geographic information

systems (GIS) have given researchers the capacity to analyze multiple layers of physical and social data in

new ways. This has led to the creation of specialized software - such as Integrated Valuation of

Environmental Services and Tradeoffs (InVEST) - that tabulates the ecosystem services generated from
9.
specific land units (Nelson et al. 2009, Bai et al. 2011, Fisher et al. 2011). The problem with this approach

is that researchers are more likely to focus on the biophysical conditions of the site without examining

how specific users may interact with or benefit from ecosystems. This follows the mainstream research

sequence in the field (Fig 1.1). Instead, research on ecosystem services would benefit from focusing on

the social and economic questions first (Cowling et al. 2008). Re-orienting the sequence of how

ecosystem services are studied, with a focus on the demand for ecosystem services first, is more likely to

capture the attention of decision makers. Moreover, a comprehensive valuation is often not necessary, as

long a research can show how policy makers can reach their objectives (Chapter 4, Powers et al. 2005,

Sangenberg and Settele 2010).

In the chapters that follow I examine ecosystem services that influence water supplies.

Throughout the dissertation, I examine how resource managers balance technological choices and

ecosystem based solutions when managing their water resources.

1.2 Study Area


Water Supply and Treatment

Barcelona depends on the waters from the Llobregat River for municipal, agricultural and

industrial uses. The Llobregat River provides the region with 45% of its drinking water supply

(Mujeriego 2006). The other half comes from either groundwater sources or the Ter watershed to the

north-east. In a normal year, the Ter-Llobregat system supplies the Barcelona Metropolitan Region with a

total of 300 hm3, two thirds for urban use, and one third for industrial use (Saur 2003).

Barcelona had historically relied on groundwater to meet its drinking water needs. However

starting in 1955, the private firm Aguas de Barcelona (AGBAR) financed the construction of the new

treatment facility on the shores of the Llobregat River at the base of the watershed (Bolaos 2004). Yet

even after the construction of the new facility, demand continued to rise, and in the 1960s, Barcelona

expanded its territorial reach for freshwater by pumping in supplies from the Ter watershed.

10.
The region of Catalonia in Spain relies largely on reservoirs to meet its water needs. Indeed,

Spain is a global leader in the construction of dams, with one of the greatest number of dams per capita in

the world (Arrojo and Naredo 1997, Bakker 2002, Arrojo 2003). In the 1950s, 1960s and 1970s, the

Spanish government built several large dams throughout the Pyrenees Mountains. Today, three major

dams regulate water flows in the Llobregat watershed: Sant Pon (1956), Baells (1976), and Llosa del

Cavall (1997). The Llobregat reservoir system has a capacity of 139 hm3 while the neighboring Ter

watershed has a storage capacity of 402 hm3.

In 1980 a consortium of private firms built a second major water treatment facility on the

Llobregat River in the town of Abrera, upstream of ABGARs existing facility. Ownership of the

treatment plant changed hands several times until the Catalan government finally took control from

AGBAR (personal comment Antoni Bernal, January 2010). The public water agency Aiges Ter

Llobregat (ATLL) now manages this facility. ATLL is a water wholesaler that treats water sources and

sells potable water to municipal providers. ATLL is publically run however the Catalan Government has

plans to outsource the management to a private firm (Expansin 2011). Both water treatment plants must

report to the Catalan Water Agency, or Agncia Catalana de lAigua (ACA).

Water managers at the ATLL and AGBAR treatment facilities have been dealing with

contaminants in the Llobregat River for decades (Lloret 2004, Estevan and Prat 2006). In particular, mine

tailings upstream release salts and bromides into the river. For years, relocating the mine tailings has been

considered financially prohibitive (ACA 2006). To alleviate the pollution coming from the mines, in

1989, ACA financed the construction of a brine collector that significantly reduced the tailings impact on

water quality. However bromides in the water supply still react with disinfectants during the treatment

process to form carcinogenic compounds called trihalomethanes (THMs) (Sorlini and Collivignarelli

2005). Drinking water standards require total THM content to be under 100 g/L as of January 1, 2009

(Royal Decree 2003).

11.
Between 2008 and 2009 both ATLL and AGBAR installed new desalination or membrane

technologies to remove contaminants and safely comply with new drinking water quality standards. The

public water treatment facility managed by ATLL installed an electrodialysis reversal (EDR) system,

while further downstream, the private water company AGBAR installed reverse osmosis (RO)

technology. The AGBAR treatment facility also extracts groundwater from an aquifer that is experiencing

seawater intrusion, further compounding the salinity problem in its water supply.

Llobregat Watershed

The Llobregat River flows 170 kilometers from the Pyrenees Mountains to the Mediterranean Sea

in the north eastern region of Catalonia, Spain (Fig 1.3). The watershed covers 4,948 km2 making it

Catalonias largest watershed entirely within the jurisdiction of the autonomous community (Lloret 2004).

The Llobregat is a fourth order stream whose primary tributaries include the Cardener, the Anoia, Rub,

and Merls streams.

The upper segment of the watershed receives 1,000 mm of precipitation a year, while the mid

section only 400 mm and the coast approximately 550 mm per year. In one year, approximately 660

cubic hectometers (hm3) of water flows down the Llobregat (Mujeriego 2006), although this discharge is

highly variable, and annual flows vary between 230 to 1,870 hm3. The river also feeds into a

groundwater aquifer below the Llobregat Delta, contributing approximately 57 hm3 per year. The rivers

hydrological contribution to the aquifer is nearly three times the contribution received from precipitation

(20 hm3/year) (Mujeriego 2006).

12.
Figure 1.3 The Llobregat watershed in north eastern Spain.

13.
In 2000 the European Union approved legislation designed to restore rivers and lakes across the

continent. The Water Framework Directive (WFD) set targets and timetables for restoring water bodies

to a good ecological status by 2015 (EU 2000). A major legislative innovation included the requirement

to conduct an economic analysis prior to major policy decisions. Non-compliance with the WFD may

result in monetary fines and reduce the member states bargaining power within the European Union.

Thus the WFD is seen as an integrative piece of legislation that responds to the expectations of watershed

planners and has raised the hopes of environmental advocates.

While the WFD marks a major turning point for freshwater management in Europe, countries are

allowed to exclude certain water bodies from meeting the good ecological status requirement if they

can demonstrate that the costs disproportionately exceed the benefits of restoration (Hanley and Black

2006). As written, this provision pushes the economic analysis to center stage, and yet the environmental

community is concerned that it will allow governments to shirk their restoration responsibility for

restoring aquatic ecosystems (personal comment Daniel Barb, October 2009). Still, the Catalan Water

Agency has worked earnestly to meet the WFD timetable and targets. As of 2010, Catalonia was the only

autonomous community in Spain to have fulfilled its obligations in the WFD before the European Union

(personal comment Pedro Arrojo, May 2010).

However Spains current budgetary crisis threatens to chip away at the progress made. Drastic

budget cuts at the Catalan Water Agency are likely to postpone restoration projects for rivers, lakes, and

estuaries. Currently the water agency is burdened with a debt of 1.3 billion that the Catalan Government

has promised to restructure and reduce (Arbol 2011, Zann 2011). Up to 40% of the staff at the water

agency may be cut by the end of 2012 (Garriga 2011). To further raise revenue, the Catalan government

is exploring the possibility of leasing its public water treatment facilities - currently managed by the

public agency ATLL - to a private firm (Garriga 2011). The prospect of transferring water treatment and

distribution activities to the private sector has raised criticisms from civil society organizations, who

14.
assert that private operators will prioritize profit over service quality, capital investment and ecological

restoration (Garriga 2011b). Critics also question why water rates did not increase in the period when the

public enterprise was entering into debt. Instead, rate increases are likely to be pushed onto the public

once the water concession is contracted with the private sector (LEconomic 2011). Negotiations are still

underway between the water agency and prospective private operators. Throughout 2012 and beyond, the

organization and structure of the water sector in Catalonia will continue to evolve.

15.
- THIS PAGE IS INTENTIONALLY LEFT BLANK -

16.
CHAPTER 2. ENVIRONMENTAL HISTORY OF THE LLOBREGAT
WATERSHED

Abstract

This chapter examines the history of water use in the Llobregat watershed to explain why urban

planners in Barcelona, Spain were compelled to install desalination water treatment technology along the

Llobregat River instead of exploring other options, including the management of ecosystem services. In

later chapters I explore how these technologies have created new opportunities for managing and valuint

ecosystem functions and structures. But first I examine how we got here -- to a high tech system of water

treatment and distribution. The answer can be found by reviewing the origins of a river pollution conflict

that has vexed water managers for nearly a century. This historical analysis also explains why water

managers have frequently preferred technological solutions over ecosystem management practices. While

ecosystem restoration alone might not have resolved the salinity conflict that confronted water managers

in the 1920s and 1930s, the outcome of that confrontation shows a clear preference for the construction of

public works. The key stakeholders in this quarrel set aside their differences in other respects and came

together to agree that the river pollution problem could be easily resolved with the construction of a brine

collector and a dam in the mid section of the watershed. The consensus on this issue set the city of

Barcelona down a trajectory of river and water management that favored structural solutions to address

their water conundrums.

This chapter is divided into two parts. In the first segment I review the salinity conflict in the

Llobregat and Cardener Rivers in the first half of the twentieth century. Previous research had identified

the origin of the conflict in 1926, but only mentioned this date in passing without providing additional

details. I have sought to take a closer look at the first decade of this conflict and study the planning

process that followed once it became clear that Barcelonas water supply was in danger of becoming

irreversibly polluted. I find that the deliberations of the CESALL Commission in 1931 and 1932 were the

17.
first exercise in watershed planning in the Llobregat basin and their recommendations have had a lasting

influence on subsequent generations of water managers.

In the second segment I show how notions of ecosystem services were discussed as an alternative

strategy for flood control in the late 19th century. Thanks to the writings of a forester, Rafael Puig Valls

(1845-1920), we know that the restoration of forest ecosystems in the Pyrenees mountains became a

serious management alternative for mitigating floods in the Llobregat watershed. This ecosystem

approach serves as a precedent for contemporary discussions on ecosystem services in the Llobregat

watershed.

Keywords: Cardener River, history ecosystem services, Llobregat River, salinity pollution, watershed

planning

18.
2.1 Introduction
Before humans withdrew water from the Llobregat River to irrigate their crops, propel their mills,

or transport their waste, the river was a major source of problems; a threat and a nuisance; an obstacle and

a barrier. First and foremost, the rivers floods threatened to wash away life and property. The first

recorded flood dates back to 1143, in an event that washed away the stone bridge at Martorell constructed

by the Romans.1 Soldiers, merchants and travelers had used the bridge as a critical river crossing along

the Via Augusta; a transportation artery along the rim of the Mediterranean that connected Imperial Rome

with the Mediterranean port of Tarraco in Hispania Citerior and the silver mines in Carthago Nova in

Hispania Ulterior.2 In the years that followed this first recorded flood, between 1143 and 1900, historians

have counted 129 additional floods along the Llobregat.3 All were large enough to wash away bridges,

boats, people and property; and at minimum, generate sufficient wreckage so as to leave behind written

record. We can safely assume that many more floods swept through the Llobregat without leaving

historical trace.

The first concession to draw water from the Llobregat was signed by King Jaume I of Aragon and

Catalonia in 1273 and granted to the monks of Sant Cugat del Valls so that they could nurture their

crops. In 1321, the Kings son, Jaume II, granted another concession allowing river dwellers to harness

the rivers energy.4 Former inhabitants along the Llobregat River have left us with clues about how they

administered the rivers waters, such as irrigation canals, or medieval bridges. Past water users have also

left us with the nomenclature of familiar places that can easily be traced back to the river; the town of

Molins de Rei was named after the first hydro-powered mill constructed with consent from the King, and

1
Codina, J. 1971. El delta del Llobregat i Barcelona. Gneres i formes de vida dels segles XVI al XX. Hores de
Catalunya. Esplugues de Llobregat: Edicions Ariel.
2
Hughes, R. 1992. Barcelona. New York: Knopf.
3
Codina, J. 1971. El delta del Llobregat i Barcelona. Gneres i formes de vida dels segles XVI al XX. Hores de
Catalunya. Esplugues de Llobregat: Edicions Ariel.
4
Ibid.
19.
Sant Andreu de la Barca was the site where one could cross the river in a boat.5 However more subtly,

and perhaps more importantly, past water users have left us with certain ideas about how rivers should be

managed. These inherited ideas configure and constrain our relationship with the waters that rivers carry.

We are all subjected to this historic influence. It is an influence that often goes unnoticed and

unquestioned. These ideas shape how we speak about rivers, our assumptions about what should be done

to them, our view of the rivers natural conditions, and our expectations about future possibilities. In this

chapter I argue that our current system of water treatment and distribution reflects a historic tradition that

has favored technological solutions to water problems. We see this in the recommendations of the

CESALL Commission in 1932, which outlined a structural approach for addressing the Llobregat Rivers

water pollution problems.

At the same time, the historic review uncovers an alternative perspective as well. Not everyone

advocated for engineered approaches to water management. This alternative was most clearly articulated

by a forester, Rafael Puig Valls, who advocated for the management of forest ecosystems in order to

mitigate flood damage. While Puig Valls made innovative proposals for his time, he was not the first to

stress human reliance on ecosystems. The ancient Greek philosopher Plato has been credited with first

appreciating humans dependency on ecosystems when he observed that the deforestation of the Attica

caused fertile soils to erode and springs to disappear.6 But it was not until the nineteenth century that

George Perkins Marsh systematically compiled evidence showing that human societies depended on

environmental support systems. Relying mostly on historical texts published in Europe, Marsh wrote Man

and Nature in 1864, which was subsequently revised as The Earth as Modified by Human Action in 1874.

Both texts synthesize the ecological knowledge of the time as it pertained to the progress (or decline) of

civilizations and human well being. Working at the nexus between ecology and human history, Marsh

5
Codina, J. 1971. El delta del Llobregat i Barcelona. Gneres i formes de vida dels segles XVI al XX. Hores de
Catalunya. Esplugues de Llobregat: Edicions Ariel.
6
Mooney, H., and Ehrlich, P. 1997. Ecosystem Services: A Fragmentary History. In G. C. Daily, Nature's Services:
Societal dependence on natural ecosystems (pp. 11-19). Washington, D.C: Island Press.
20.
wanted his description of ecological processes to excite an interest in a topic of much economical

importance.7

Marsh lamented that human societies appreciated ecosystem functions only after their services

were lost. He wrote, the destructive agency of man becomes more and more energetic and unsparing as

he advances in civilization, until the impoverishment, with which his exhaustion of the natural resources

of the soil is threatening him, at last awakens him to the necessity of preserving what is left, if not

restoring what has wantonly been wasted.8 Marsh provides examples of environmental mismanagement

in Asia Minor, Northern Africa, Greece and Alpine Europe, whereby action by man has brought the face

of the earth to a desolation almost as complete as the moon.9 After reviewing countless cases of decline

across different cultures and continents, Marsh became convinced that unwise ecosystem management

contributed to societal impoverishment. The decay of these once flourishing countries is partly due, no

doubt to the direct violence of human force..[and] ignorant disregard of the laws of nature.10 Marsh

noted that ecosystems retained valuable soils and stabilized hydrologic flows. Later I will discuss Marshs

observations concerning hydrology in more detail, but for now, suffice to say that Marsh was probably the

first to write about human dependence on functioning ecosystems with such detail.

However before embarking on the historic analysis, I begin with the present; where we find that

Barcelona has built a system of water production that is technologically sophisticated and energy

intensive. Taking a brief look at water distribution today gives us clues about the sorts of ideas that water

managers have inherited from the past.

The high tech water treatment systems installed along the Llobregat River are the latest in a series

of technological fixes to a river pollution problem that is decades old. The salinity contamination of the

Llobregat from mining activity has even been called Catalonias most severe environmental problem; and

7
Marsh, G. P. 1874. The Earth as Modified by Human Action. A New Edition of Man and Nature. London:
Sampson Low, Marston Low, and Searle. Republished by Elibron Classics 2006. p 9
8
Ibid. p 39
9
Ibid. p 43
10
Ibid. p 5
21.
if the new desalination systems serve as evidence, the issue remains unresolved. The history of salt

pollution in the Llobregat and Cardener Rivers helps us understand the connections between past

conflicts, the solutions put forward, and contemporary water challenges. I trace this river pollution

conflict from its origin in 1926, through the Spanish Civil War 1936-1939, and then rapidly bring it up to

the present.

Then I move on to discuss the competing proposals to mitigate flood damage in the Llobregat

watershed, with a focus on the Cabrianes dam project proposed in the 1890s. In this discussion on flood

management we can see a clear alternative to the mainstream approach that relied on structural solutions.

The writings of Rafael Puig Valls advocated for the management and restoration of ecosystem services

for the benefit of downstream water users. Puig Valls sought to apply knowledge from the ecological

sciences to construct a viable alternative to the structural approaches to flood control that dominated the

era. Interestingly, the debate that Puig Valls began in the nineteenth century closely mirrors current

ideological divisions in the watershed today. Many of Puig Valls proposals for reforestation were

eventually adopted during the Second Spanish Republic (1931-1936), but ultimately his ideas about

ecosystem management failed to displace the hegemony of structural approaches to water management.

And by taking a look at contemporary water treatment in Barcelona, we can appreciate just how

technologically intense their system of water distribution has become.

2.2 Desalination of the Llobregat River


The Llobregat River is a freshwater system that has become brackish due to mining activity

upstream. To remove the salts and dissolved solids from the river, water managers have turned to

sophisticated desalination systems. In the late 2000s, the combined cost of installing both desalination

22.
systems surpassed 130 million.11 Moreover the operation of these treatment technologies requires an

enormous amount of energy. The electrodialysis reversal (EDR) system run by Aiges Ter Llobregat

(ATLL) consumes approximately 13 Giga Watt hours (GWh) per year, at a cost of 1.2 million per year.

Downstream, the ultrafiltration and reverse osmosis system operated by Aguas de Barcelona (AGBAR)

consumes approximately 21 GWh per year, for an energy expense of 2 million per year.12 In other

words, the energy consumed to treat water from Llobregat River each year would be sufficient to meet the

energy needs for the entire city of Barcelona; including gas for cars, electricity for air conditioners, lights

and televisions etc... for two full days. The decision to install desalination technology on the banks of the

Llobregat River was a strategic choice to trade energy for freshwater. It also committed the Barcelona

metropolitan region to spend 3.8 million per year to desalinate water that, at its source in the Pyrenees, is

of superb quality.

Not only is desalination expensive and energy consuming, but the membranes are complex to

operate and vulnerable to fouling. Feed water entering the membranes must be groomed to meet exacting

pre-treatment standards. Feeding the high-tech apparatus with unbalanced water could severely damage

the capital investment and reduce long term efficiency. For example, aluminum compounds are used at

the beginning of the treatment process to remove suspended solids and particles from the river water.

However the aluminum added must be removed before entering the reverse osmosis membranes to

prevent fouling. Water treatment engineers precipitate the aluminum by reducing the pH, but excessive

pH reduction will cause the water to release carbon dioxide which is later added, at a cost, for more

11
Valero, F., and Arbs, R. 2010. Desalination of brackish river water using Electrodialysis Reversal (EDR) Control
of the THMs formation in the Barcelona (NE Spain) area. Desalination 253:170-174.; Pedraz Yaez, G. 2007.
Proyecto y obra de la Mejora del Tratamiento de Aguas por Osmosis Inversa en la ETAP de Sant Joan Despi.
Documento 1. Memoria. Barcelona: Aguas de Barcelona.
12
Calculations by author
23.
balanced smell and taste.13 Any misstep in the treatment process could drastically increase treatment

costs or damage the expensive membranes.

What led Barcelona down this path of technological sophistication, energy intensity and high

treatment costs? This historical review will analyze the circumstances that led water managers to favor

this complex system for water treatment, over alternative options of ecosystem management.

13
Pedraz Yaez, G. 2007. Proyecto y obra de la Mejora del Tratamiento de Aguas por Osmosis Inversa en la ETAP
de Sant Joan Despi. Documento 1. Memoria. Aguas de Barcelona: Barcelona.
24.
Figure 2.1 The Llobregat Watershed

25.
2.3 Salinity Conflict
Minerals have been extracted from Llobregat watershed for centuries, especially in the mountains

surrounding the medieval town of Cardona (Fig 2.1). The Romans who settled in Cardona extracted

sodium chloride (NaCl), more commonly referred to as table salt, from an emerging salt dome that

penetrated the earths surface.14 Geologists call these salty protrusions diapirs, and they are part of the

areas natural geology. The salt deposits are the product of tectonic movements that pushed the Iberian

Peninsula into mainland Europe.15 The collision enclosed ocean waters in an enormous lake filled with

salt water. Over time, the water evaporated, leaving behind layers of minerals including sodium chloride,

potassium chloride, and magnesium chloride. Further tectonic movements buried the lake bed, and then

pushed certain sections back up to the surface. With the exception of the salt diapir emerging in Cardona,

the salts are usually hundreds of meters below ground. In Sria, where the salt minerals are closest to the

surface, the salt is 150 meters below.16 Mining for table salt peaked in the 18th century and then declined

steadily. The extraction of salt was replaced by a more valuable mineral: potassium chloride (KCl).

Local legend asserts that potassium chloride was discovered by accident when locals noticed that their

livestock refused to eat the salts they would normally devour.17 Laboratory analysis by the Frenchman

Ren Marcay in Bordeaux confirmed the potassium find in 1912.18

News of the discovery generated considerable excitement. Potassium was a valuable fertilizer for

agriculture, and useful as an explosive in war. Prior to the confirmation of potassium deposits in Sria, it

was believed that only Germany, France and Poland had domestic sources.19 Under pressure by regional

14
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
15
Arnau Reitg, R. 1981. La mineria del Bages: una visi retrospectiva. XXVI Assemblea Intercomarcal d'estudiosos
a Manresa (pp. 53-58). Manresa: Centre d'Estudis del Bages.
16
CESALL Report.1932. Arxiu Nacional de Catalunya (ANC) Fons 547. Agncia Catalana de l'Aigua. Unitat 479.
17
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
18
Ibid.
19
Arnau Reitg, R. 1981. La mineria del Bages: una visi retrospectiva. XXVI Assemblea Intercomarcal d'estudiosos
a Manresa (pp. 53-58). Manresa: Centre d'Estudis del Bages.
26.
politicians in Manresa, the Spanish Cortes in Madrid quickly modified the mining legislation to expedite

the extraction and commercialization of potassium.20

The Belgium chemical firm Solvay took the lead in developing the potassium mines in the

Llobregat watershed. They began their operations in Sria - where the initial discovery came to light -

located on the Cardener River between Manresa and Cardona (Fig 2.1). Solvay was a reputable chemical

firm founded by the Belgian chemist Ernest Solvay (1838-1922), and in 1920 they set up their Spanish

subsidiary Minas de Potasa de Sria. That year a managerial elite from Belgium descended onto the

small town to run the operations. Solvay sent Norbert Fonthier, a disciplined ex-Army officer and World

War I veteran to run the mining subsidiary.21 Fonthier was a highly disciplined manager, who paid keen

attention to detail, and dutifully relayed major decisions back to headquarters in Brussels.

As an initial sign of goodwill to their Spanish hosts, in 1920 Solvay donated funds to construct a

public water system that would guarantee 150,000 liters of potable freshwater per day for the towns

3,200 residents.22 In a display of appreciation for this generous gift, Sria renamed the street in front of

City Hall, Avenida Ernest Solvay, making the name of the Belgian chemist ubiquitous on municipal

letterhead. The town also engraved two plaques to commemorate the privately financed water system; one

on the faade of City Hall and a second in the City Hall courtyard (Fig 2.2). What the residents of Sria

could not fathom was that the same company that brought them clean water to their town of 3,200 would

soon threaten the water supply of one million residents in Barcelona.23 It is hard to believe that such a gift

was coincidental, and suggests that the Belgian chemists and engineers already understood the

relationship between their plans to extract potassium and water pollution. Perhaps because Minas de

20
See letters between the Mayor of Manresa, Sr. D. Maurici Fius Pala and the Director of Agriculture, Mines and
Mountains in 1914. Arxiu Comarcal del Bages, Manresa.
21
The first Director of Minas de Potasa de Sria was Louis Dupont, who was followed by Van Styovoort, and
finally Norbert Fonthier who led the firm until the Spanish Civil War.
22
Reguant, J. 1997. Sria: Histria en imatges. 1894-1975. Angle Editorial. Manresa, Spain.
23
Tello, E., and Ostos, J. 2011. Water consumption in Barcelona and its regional environmental imprint: a long-term
history (1717-2008). Regional Environmental Change 1-15.
27.
Potasa de Sria anticipated their companys huge demand for water, they first ensured that local needs

were met, and resistance averted.

Figure 2.2 A plaque in Sria hangs on the faade of City Hall that that commemorates the renaming of
the street Avenue Ernest Solvay. The plaque reads: Avenue Mr. Ernest Solvay. Illustrious chemist and
engineer, maker of good works for humanity and founder of the Society Solvay who donated potable
water to Sria. 1838-1922 (Photograph by the author)

Small scale potassium mining in Sria had begun in 1918, but poor road conditions to and from

the mines severely limited the volumes they could haul to market each day. To scale up production,

Minas de Potasa de Sria built a rail line connecting their mine to the port of Barcelona where their

minerals could then be shipped around the world. With help from the Spanish authorities, the rail line

was finished in 1925, after which, large scale mining began. The new rail line allowed freight cars to haul

potassium out from the mountains near Sria, and the rise in sales was meteoric. The mining company

sold 2,989 tons of potassium chloride in 1925, a figure which more than doubled the next year to 8,386

tons, and doubled again to 16,084 tons in 1927. By 1928 operations were in full swing and Minas de

Potasa de Sria employed 857 workers.24 That same year, the Spanish Government decreed a law

24
Reguant, J. 1997. Sria: Histria en imatges. 1894-1975. Angle Editorial. Manresa, Spain.
28.
obliging the mining company to produce a minimum of 30,000 tons for the Spanish market. Exports were

only permitted if internal demand was satisfied, and the export price had to be higher than domestic

prices.25

Minas de Potasa de Sria had close ties with the Spanish government and monarchy. These

relations were developed with the help of Don Amadeu Hurtado de Glvez-Caero, hired by the mining

company to represent the firm in Madrid and to make the appropriate government contacts when needed.

As an example of these close ties, in 1924 a mining executive from Minas de Potasa de Sria was

knighted to the Order of Queen Isabella la Catolica. But the mining companys close relationship with

the Spanish government weakened during Spains transition to a democracy in 1931, after which, more

transparent institutions were established to manage the rivers waters.

Water Pollution Conflict

Doa Antnia Burs Borrs (1854-1928) was the first to complain about the salt being dumped in

the Cardener River by Minas de Potasa de Suria (Fig 2.3). She was a woman with extraordinary wealth,

power and personality, and yet her impressive pedigree was insufficient to halt the river contamination.

Clashes over water rights were nothing new for the Burs family. Antnia Burs came from a powerful

industrial class that had made their fortune from Catalonias waterways, and were highly sensitive to

modifications in the rivers flow. For decades already, these industrial families had fought to control the

rivers current and the energy it helped them produce. Therefore the appearance of the mining company

in the 1920s threatened to break the existing equilibrium among industrial water users. However in the

legal complaint Burs filed against the mining company in 1926, she added another new dimension to this

tussle water quality. In addition to her claim against the illegal water withdrawal, she asserted that

Minas de Potasa de Sria was impurifying the rivers waters. The mineralized effluent from the mines

25
Gazeta de Madrid 1928. Nm 15. 15 enero 1928 p 436.
29.
corroded the metallic turbines that propelled her hydro-powered textile factory in Antius.26 Her claim

against the mining company, one that combined charges about water withdrawals and water quality,

makes it difficult to ascertain the relative weight she placed on water quality considerations. Other

historians have suggested that her charge concerning water pollution was merely a way to pile on

accusations against the mining company, when in practice, the real conflict was about protecting water

rights.27

Certainly, losing water rights was an expensive proposition. Her business relied on generous

flows from the Cardener River to produce valuable electricity. Antnia Burs owned a water concession

for 5 cubic meters per second that she used to propel a hydro-electric turbine in Antius with a 7.2 meter

vertical drop (Fig 2.4). A hydro-generator with these characteristics could produce 212 kW of

electricity.28 When flows were too low, as often was the case in summer months, the factory was forced

to switch to steam engines. Of course, factory owners were reluctantly to turn on the steam engines,

because of the cost of buying and burning coal.

Antnia Burs Borrs was born into a family of the emerging industrial bourgeoisie that had

relied on hydropower for decades. Antnias father, Esteve Burs Arderiu, had founded a large company

town that produced fabric by capturing hydraulic energy from the Llobregat River (Fig 2.5). Hydropower

was a relatively inexpensive energy source, although not without its inconveniences. River banks were

rocky, isolated from towns, had poor road access and were vulnerable to periodic flooding. These desolate

sites lacked even the most basic infrastructure. The distance between the planned industry and existing

towns presented a considerable challenge for prospective entrepreneurs. The factory workers would need

more than a place to work, but also a place to live, to house their families and to educate their children.

26
Personal Communication. Esteve Torrens Prez de los Cabos. January 2011.
27
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
28
The electricity generated is a function of head (vertical drop), discharge and gravity. Therefore the electrical
power in Watts generated under these circumstances is: 7.2 m*5,000 liters/s * 9.8 m/s2 = 353 kW. I assume 60%
efficiency so only 212 kW of electricity is generated. For more information on the calculation of electrical
generation with hydropower see http://www.reuk.co.uk/Calculation-of-Hydro-Power.htm
30.
Industry leaders responded by embedding their textile mills within a company town complete with

employee housing, a school, a general store, a church, and often a theater. In effect, the industrialists who

set up shop along the Catalonias rivers accepted their workers personal and social needs as a cost of

business. Their business model went beyond the creation of an efficient factory floor, and attempted to

construct an efficient social structure for production. They created isolated societies in which everything

revolved around the production of textiles. The patrons paternalistic provision for their workers and

families earned the owners a degree of control over their labor force in addition to loyalties that lasted

generations. In exchange, the workers received a secure income and above average working and living

conditions for the era.

Antnia Burs married Llogarri Torrens Serra, whose family had also made a fortune producing

textiles. Llogarri Torrens died in 1915 leaving her a widow. Over time, she became a well-known

resident of Manresa and locals began referring to her extravagant home as La Buresa, a name that remains

today. Antnia Burs left behind a paper trail that suggests that she was an active widow who managed

the properties of her late husband. One of these properties included the Antius textile mill on the

Cardener River; the site where the water pollution conflict along the Cardener and Llobregat rivers

originally began.

By 1925, mining activity had begun in Sria in earnest. As mining activity increased, so did the

volume of salty effluent being dumped into the Cardener River by the mining companies. While it is

likely that several farmers felt the impact of the salty waters in their irrigation canals, few had the political

clout to challenge the mining company. Antnia Burs became the first to break the silence. In a letter

dated September 30th 1926, she filed a complaint with the Water Resources Authority against the mining

company Minas de Potasa de Sria S. A. for impurifying the waters of the Cardener.29 In an era when

woman were excluded from business and legal affairs, it was remarkable that Antnia Burs was the first

29
Letter from Delegat of Serveis Hidrulics del Prineu Oriental to Mayor of Barcelona. 7 October 1933. Obres
Pbliques. CV 50/237 1933. La contaminaci de les aiges del riu Llobregat amb motiu de la explotaci de les
mines de potasa de Sria. Arxiu Municipal Contemporani de Barcelona.
31.
with the courage to raise the issue. Although she died soon after her complaint was filed, her son,

Esteban Torrens Burs, continued to organize the industrial water users and their oposition to the illegal

dumping of saline waters into the Cardener River.

Antnia Burs inaugurated a water pollution conflict that would last for decades. The subsequent

controversy would transform water management, infrastructure development and water treatment in the

Barcelona metropolitan region for the next seventy five years and well into the twenty first century. As

such, her letter to the Water Resources Authority marked the opening chapter in an enduring conflict over

salinity values in the Cardener and Llobregat Rivers. More powerful figures would eventually join her

cause, including the private water company Sociedad General Aguas de Barcelona, the Barcelonas

Municipal Laboratory, the La Vanguardia newspaper, and the Mayor of Barcelona. Still, being first did

not help her case. To the contrary, she would also be the first to fall in defeat against the powerful mining

interests that repeatedly succeeded in evading responsibility for their effluents. But her claim did help

justify the formation of a commission to study the salinity problem in the Cardener and Llobregat Rivers

in 1928, another in 1929 and a third investigation in 1931. However by the time the last and most

influential investigation began in 1931, the potash mines had already consolidated their power in

Catalonia.

Figure 2.3 Antnia Burs Borrs (1854-1928). (Photograph courtesy of Esteban Torrens Prez de los
Cabos)
32.
Figure 2.4 The Antius colony along the Cardener River owned and managed by the Torrens Burs
family. (Photograph courtesy of Esteban Torrens Prez de los Cabos)

Figure 2.5 The Llobregat River in its passage through Castellbell i el Vilar. The Burs factory and
company town is on the left bank of the river, and was built by Esteve Burs Arderiu, father of Antonia
Burs Borrs. (Photograph by the author)

33.
Figure 2.6 The modernist home owned by Antonia Burs Borrs in Barcelona on Ausias March 46.
Notice the inscription of initials A.B. above the doorway. (Photograph by the author)

Watershed Planning

In January 1927 the Governor of the Province of Barcelona reviewed the lawsuit submitted by

Burs and ordered Minas de Potasa de Sria to halt the illegal water withdrawals until the company had

obtained the appropriate water concessions. The mining company challenged the Governors resolution

and a legal battle ensued between the extractive industry and the owners of the textile factories. The

plaintiffs were organized by the son of Antonia Burs, Esteban Torrens Burs, who rallied together an

alliance of industry leaders to defend their pre-existing water rights. New diversions upstream, especially

illegal ones from a growing industry, threatened their energy source and posed a significant challenge to

their historic control over the Llobregats main tributary.

34.
With tensions rising between Minas de Potasa de Sria and the industrial leaders along the

Cardener, the Spanish Ministry for Economic Development dispatched their lead engineer, Carlos Santa

Mara, to review the claims and draft a summary report with recommendations. This would be the first of

several investigations. On March 31st 1928, Santa Mara concluded that Minas de Potasa de Sria was

indeed, illegally diverting water from the Cardener River, and that the mining process appreciably

increases the mineralization of the river. While he did not feel that this contamination was an immediate

danger to public health, he did recommend for the formation of a committee comprised of specialists

representing a wide range of backgrounds to study the urgent issue.30 He also suggested that Minas de

Potasa de Sria install devices to monitor the quality and volumes of their effluent.

None of these recommendations materialized. Instead, on November 20th 1928, the Spanish

Government requested that experts from the mining industry take charge of the issue. This group of

mining engineers and geologists were organized in a governmental panel called the Superior Board on the

Extraction of Potassium Salts.31 The Spanish government asked them to determine the maximum

admissible amount of salt that could be dumped into the river. At this point, there was no mention of how

the mining effluents were impacting drinking water supplies. The conflict remained between the mining

company and an elite group of industry leaders.

The Superior Board on the Extraction of Potassium Salts elected Agustn Marn y Bertran de Lis

to lead the investigation into the salty effluents flowing out of Sria. He was asked to coordinate a team

that would collect new data and make recommendations about how to harmonize industrial processes.

The Superior Board on the Extraction of Potassium Salts was comprised of mining specialists; and as a

group, there were exceptionally well qualified to study the mines impact on the Cardener and Llobregat

Rivers. More than anyone, they had an intimate understanding of the methods used by the mines to

extract potassium. They knew, for example, that Minas de Potasa de Sria depended on large volumes of

30
CESALL Report.1932. Arxiu Nacional de Catalunya (ANC) Fons 547. Agncia Catalana de l'Aigua. Unitat 479.
31
Junta Superior de Explotacin de Sales Potsicas
35.
water to dissolve the potassium chloride in the first phase of extraction. Once dissolved into a liquid

porridge, the miners separated the valuable potassium from other worthless compounds through a

decanting process. The problem was how to dispose of the remaining fluids, saturated in sodium chloride

and magnesium chloride. The easiest option was to open the flood gates leading into the Cardener River.

So when Agustn Marn was asked to study the salinity problem associated with the mining industry in

Sria, there was no doubt in his mind that the activity was releasing large volumes of chloride into the

Cardener River. The critical question was how much.

Agustn Marn set out to answer this question by taking water samples from the Cardener River

above and below the mines. The samples taken in May 1929 showed that the mining company dumped

approximately 26,940 tons of sodium chloride and 51,840 tons of magnesium chloride into the river each

day (Table 2.1). Water quality samples were taken again in August 1929 with even more dramatic results

-- the volume of magnesium chloride quadrupled downstream of Minas de Potasa de Sria and sodium

chloride increased 50%. The Superior Board on the Extraction of Potassium Salts concluded that the

mines dumped approximately 100 tons of chloride into the river per day, diluted in a daily effluent of

15,000 cubic meters.32 However their internal discussions remained private, and the report they began to

prepare for the Spanish Government probably would have stayed confidential had the problem not

escalated. 33

While the mining experts were still tabulating the chloride loads released into the Cardener River,

the consequences of the chloride effluents began to be felt downstream. Sufficient chloride had travelled

32
CESALL Report 1932. Arxiu Nacional de Catalunya (ANC). Fons 547. Agncia Catalana de l'Aigua. Unitat 479
33
Marn co-lead the first investigation with Alonso Martnez. Mr. La Rosa and Mr. Santa Mara travelled to the
Cardener to take water quality samples, and Marn and Martnez were responsible for presenting the results to the
Board. Once the Board was informed that Sociedad General Aguas de Barcelona has begun legal proceedings
against the mining company, the Board was approached by the representative of the mining company, Glvez-
Caero for assistance, at the Board meeting on Christmas Eve 1930. The Board then asked Mr. Kindeln and Mr.
La Rosa to visit the water company and the groundwater pumping stations in early 1931. Mr. Menndez Puget was
asked to visit the mining companies. Once the CESALL commission was created in March 1931, these working
groups were dissolved and they passed on their information to Marn, who sat on the CESALL commission.
(CESALL 1932)
36.
down the Cardener River, into the main stem of the Llobregat, and then percolated into the aquifer

beneath the Llobregat Delta, that chemists had begun to detect a rise in chloride concentrations in the

groundwater wells starting in 1926. In Molins de Rei, located several miles upstream of Barcelonas

wells, chemists noticed that chloride concentrations jumped 60% between 1928 and 1930; from 137 mg/L

to 220.4 mg/L. 34

Date Site MgCl2 MgCl2 MgCl2 NaCl NaCl NaCl River


(mg/L) (Kg/day) released (mg/L) (Kg/day) released to Discharge
to river river (m3/hour)
(Kg/day) (Kg/day)

6 May 1929 Upstream 39 18,720 146 70,080


Downstream 147 70,560 51,840 204 96,920 26,940 20,000

10 August 1929 Upstream 123 27,457 213 47,544

Downstream 1,110 247,732 220,296 390 87,048 39,501 9,300

7 November 1929 Upstream 110 18,480 213 35,784

Downstream 279 46,872 28,372 1,200 201,600 165,640 7,000

13 December 1929 Upstream 100 24,000 198 47,520

Downstream 359 86,160 62,160 534 128,160 80,640 10,000

25 March 1930
Upstream 69 57,760 69 57,960

Downstream 129 108,360 50,400 164 137,760 79,800 35,000


Means excluding the highest two values 40,108 56,713

Table 2.1. Results from the first water quality samples taken to assess the impact of mining effluents on
the salinity of the Cardener River between May 1929 and March 1930 by the Superior Board on the
Extraction Potassium Salts, upstream and downstream of mining activities in Sria. Reproduced from
data in CESALL (1932).

Simultaneously, Sociedad General Aguas de Barcelona detected similar increases in chloride

values in the freshwater aquifer at the pumping stations in Cornell (Fig 2.1). The aquifer was monitored

34
Elsewhere in the CESALL report, additional data is presented from the wells in Molins de Rei: 83.33 mg/L Cl
(August 1928), 124 mg/L Cl (July 1931), 187.7 mg/L (December 1931).
37.
closely given that it provided much of Barcelona with its drinking water. As a reference point for what

chloride concentrations were before mining activities, in 1915, three years prior to any mining of

potassium, Sociedad General Aguas de Barcelona recorded a chloride concentration of 80 mg/L in the

aquifer.35 Ten years later, in 1925, precisely the year when industrial potassium mining began in Sria,

the water company recorded a similar baseline value of 79 mg/L.36

But by late 1929, the rise in chloride values in the Llobregat aquifer became a grave concern.

Sociedad General Aguas de Barcelona gathered the evidence and presented their case to the responsible

water agencies in April 1930.37 These findings seriously concerned everyone involved, since no

technology was capable of removing the dissolved solids from the water supply. A rapid increase in

chloride concentrations could permanently contaminate the water source, causing major disruption to the

city of Barcelona, and financial ruin for the water company. Sociedad General Aguas de Barcelona

quickly began to mobilize its resources and influence in order to protect the aquifer from further

contamination.

The water company executive, Jos Mara Soler Nolla, became the outspoken point person on this

issue for Sociedad General Aguas de Barcelona. When Soler Nolla contacted the responsible water

agencies in 1930, he learned that the water company was not the first to initiate legal proceedings against

Minas de Potasa de Sria for illegally dumping their waste into the river. The industrial leaders of the

Cardener valley, led by the son of Antnia Burs, Esteban Torrens Burs, in collaboration with Luis

Argem, and the Baron of Montclar, had already filed a case. But the recommendations from the first

investigation did not go far, and the second report was still being drafted by the mining experts behind

closed doors in Madrid.

35
La Vanguardia 1935. March 23. La Salinidad de las Aguas del Llobregat. p 6.
36
CESALL Report.1932. Arxiu Nacional de Catalunya (ANC) Fons 547. Agncia Catalana de l'Aigua. Unitat 479.
37
Archive Aiges de Barcelona. Box 5615. Folder 15. Informe sobre la influencia que tienen las explotaciones
mineras en la cuenca del ro Llobregat en el aprovechamiento de aguas sublveas del mismo ro para el
abastecimiento de Barcelona y propuesta provisional. 3 November 1931.
38.
The threat of further contamination led the water company and their affiliated groundwater firm,

Empresa Concesionaria de Aguas Subterrneas del ro Llobregat, to file their first lawsuit requesting that

the Spanish Government address the salt contamination flowing from the mining town of Sria.38 The

entry of Sociedad General Aguas de Barcelona into the controversy escalated the water conflict. What

previously was an isolated dispute along the Cardener River became a regional confrontation of national

significance.

Sociedad General Aguas de Barcelona filed their claim during a transitional period between the

Primo de Rivera dictatorship, which ended in January 1930; and the democratic governments of the

Second Spanish Republic, inaugurated in April 1931.39 The transitional government took a more active

role mediating between the mining company in Sria and the water company in Barcelona.

The Spanish Government wanted a resolution that would meet the needs of both parties. It was

desirable for the mines to produce potassium fertilizers upstream, but downstream, the river needed to

arrive clean enough so that it would not harm existing users, especially municipal water suppliers. The

trouble was that the stakes were enormously high. Miscalculation could wipe out Barcelonas primary

source of freshwater. Salinization of the aquifer would also drive the water company into financial ruin.

This dismal outcome threatened to reverberate throughout the Spanish economy since Spanish banks had

recently acquired the private water company from French investors. Therefore the question regarding

how much salinity should be permitted to flow into the river was far too important to be left to the mining

interests alone to decide. More stakeholders needed to be brought to the table, and the recommendations

to emerge from the committee of experts needed to meet the interests of all involved.

38
Anonymous 1931. Informe sobre la influencia que tienen las explotaciones mineras en la cuenca del rio Llobregat
en el aprovechamiento de aguas subalveas del mismo rio para el abastecimiento de Barcelona y Propuesta
Provisional. AGBAR Archives. Box 5616. no. 15.
39
Spains Second Republic began following a landslide victory of pro-democratic political parties in Spains
Municipal Elections in April of 1931 and would last until the uprising led by General Francisco Franco in 1936.
39.
To meet these diverse needs and interests, a third investigation began in March 1931. The

Commission for the Study of the Salinity of the Waters of the Llobregat (CESALL)40 was tasked to

research the same questions as the investigations that came before them: quantify the extent of the salinity

pollution caused by the mining activity in Sria and recommend solutions to resolve the problem. The

composition of the CESALL commission struck a delicate balance among competing interests. Two

members were from the Water Resource Agency of the Eastern Pyrenees; one from the Mining

Department; another from the Public Health Department; the Director of the Municipal Laboratory of

Barcelona, Pedro Gonzalez; and finally, Agustn Marn y Bertran de Lis, representing the Superior Board

for the Extraction of Potassium Salts as the de-facto representative of the mining interests. Private parties

were invited to the CESALL commission as non-voting members: Jos Soler Nolla representing Sociedad

General Aguas de Barcelona; and Norbert Fonthier on behalf of Minas de Potasa de Sria. The Royal

Order that created the CESALL commission also specified that the expenses associated with the

commission would be financed by Minas de Potasa de Sria.

Starting in May 1931, the CESALL commission met every other week at the conference room of

the Water Resource Agency of the Eastern Pyrenees on the Via Laietana no.10 in Barcelona. Once the

CESALL commission was established, events moved quickly. By the second CESALL meeting on June

2nd 1931, the second potassium mining firm, Unin Espaola de Explosivos had begun extracting potash

along the Cardener River in Cardona. This worked to the advantage of Minas de Potasa de Sria because

it allowed them to point fingers at their neighbor for contributing to the salinization of the Cardener and

Llobregat Rivers. The new arrivals also allowed Minas de Potasa de Sria to claim that they were being

unfairly burdened with the expenses of the CESALL commission, when in practice, other mining

companies were also contributing to the problem. Within the year, a third mining company, Potasas

Ibricas, would also set up operations on the right bank of the Llobregat River near Sallent, further

adding to the mix of polluters.


40
Comisin para el Estudio de la Salinidad de las Aguas del Llobregat (CESALL)
40.
Anticipating that the salinity problems would only get worse, the CESALL got to work

immediately. They began with a fact finding mission to quantify the damages associated with the salty

effluents. The CESALL commission sent out a public inquiry asking farmers, municipal governments,

and industry leaders, to report the damages incurred by the river salinization.41 Farmers reported that their

crops had suffered, but that the specifics eluded quantification because too many variables obfuscated the

precise causal relationships between river water quality and harvest yields. Furthermore, farmers expected

that the salinization of their land would be gradual, rendering their fields barren over time. There was no

doubt in farmers minds, however, that salinity values had increased. The Municipal Government of Olesa

de Montserrat wrote on behalf of their residents in the summer of 1931, That the water in the Llobregat

is now saltier than it was a few years ago, is public knowledge, and that this fact is unfavorable to

agriculture, is evident.42

Of all the institutions who responded to the CESALL Commissions request for information

regarding the salinity of the Llobregat River, Sociedad General Aguas de Barcelona was the most

interested party. Their business depended on protecting the aquifer whose waters they treated and

pumped into the homes of the urban elite.43 The chloride entering the aquifer was irreversibly changing

the waters composition. Even small amounts could alter the taste to the palate. While the water

professionals did not foresee any health problems associated with chloride in the drinking supply, there

was a concern that the unpleasant taste would discourage new users from connecting to the urban

network, and provide an excuse to use contaminated wells instead.44

41
Boletn Oficial de la Provincia de Barcelona (BOPB) 1931. Comisin para el Estudio de la Salinidad de las Aguas
del Llobregat. 9 June 1931. No 137. Nmero disposicin 7472.
42
Letter from town of Olesa de Montserrat to CESALL in response to the public call for information on the salinity
problems in the Llobregat watershed (BOPB. 9 June 1931). AGBAR Archives. Box 5615. No. 12.
Que el agua del Llobregat, de unos aos a esta parte baja ms salada que antes, es un hecho pblico, que esta
particularidad no puede ser favorable a la agricultura, es evidente
43
Martn Pasqual, J.M. 2007. Aigua i Societat a Barcelona entre les dues Exposicions (1888-1929). Doctoral
Dissertation. Departament dHistria Contempornea i Moderna. Universitat Autnoma de Barcelona.
44
Lo que ocurre es que el agua llegara a ser salada y, por consiguiente, desagradable para la bebida, y entonces
las gentes la rechazaran. Y como la inmensa mayora de los barceloneses no puede permitirse el lujo de beber
41.
For years Sociedad General Aguas de Barcelona had relied on the filtration services of the

aquifer which removed particles and suspended solids. The director of Sociedad General Aguas de

Barcelona proudly claimed that the waters captured at our wells in Cornell, after a horizontal filtration

of various kilometers and a vertical filtration of forty meters, has a biological purity that cannot be beat by

any water source in the world. He added, But unfortunately, not even this filtration nor chemical

treatment is capable of separating the sodium chloride from the water.45

The private water company was also nervous about the destabilizing effects that this

contamination episode could have on its consumers and the working class. The political risks were

emphasized in the water companys concluding remarks to the president of the CESALL Commission:

Forgive me, Don Enrique, for my insistence, but the issue concerns me more and more,

especially during these times of quarrels and popular revolts, I do not know what would

happen if one day, after serving Barcelona with salt water, we would force people to

endure an enormous financial burden to replace the water that now flows to their faucets

without any public expense. 46

During the summer of 1931 the salinity problems were aggravated by drought. In July and August

1931, the Cardener River started to dry up, and the salt concentrations spiked. That year the flow stations

aguas minerales embotelladas, se recurrira al agua de los pozos. Y esto s que constituira un grave peligro para la
salud, pues no controlados ni vigilados estos pozos, la mayora de ellos contienen agua que no rene las
condiciones higinicas necesarias para su consumo. Interview with Dr. Pedro Gonzlez, Director of Barcelona
Municipal Laboratory. La Vanguardia 1935. Voz de Alarma. 23 March 1935.
45
La Vanguardia 1935. La Salinidad de las Aguas del Llobregat. 23 March 1935. p 6.
46
Letter from Sociedad General de Aguas de Barcelona to CESALL. 17 August 1931. AGBAR Archives. Box 5615.
No. 12.
Perdone, Don Enrique, que le moleste tanto; pero el asunto me intranquiliza cada vez ms porque en estos tiempos
de querellas y explosiones populares, no s lo que pasar si un da trs servir a Barcelona agua salda, obligamos a
sus habitantes a soportar una carga financiera enorme para poder sustituir al cabo de unos aos el agua que hoy, sin
ningn gasto pblico, mana normalmente de sus grifos.
42.
in Cardona and Martorell recorded average discharges that were half of normal.47 In response to the rising

chloride values, the Public Health Commission asked the Catalan Parliament to intervene immediately

and require the mines to reduce their discharges into the river so that the Llobregat River would not

surpass a chloride concentration of 250 mg/L.48

When the CESALL commission reconvened from summer break on September 1st 1931, the

Director of the Municipal Laboratory, Pedro Gonzlez insisted that the CESALL commission write the

Provincial Governor of Barcelona requesting that regulations be established immediately, without waiting

for the final conclusions of the CESALL investigation. The commission agreed, and within two weeks the

Governor dictated an order that chloride concentrations could not surpass 250 mg/L on the Llobregat

River at the confluence with the Cardener.49 This provisional decree would buy time for the CESALL

commission to study the issue in greater detail. The public health commission would monitor compliance

and the laboratory expenses would be paid by the plaintiff parties if the measurements were below the

limit, whereas expenses would be paid by the mining companies if a violation was found. The penalty

imposed on the mining companies for the first violation would be the closure of mining operations for one

day. A second violation would require closure for a week, and a third violation would be punished by

indefinite closure.50

The CESALL Commission released its final report in the spring of 1932. The lengthy document

was complete with tables, figures and elegant maps of the watershed. The commission concluded that

four measures were necessary to bring the salinity problem under control. First, the commission proposed
47
Plan General de Obras 1936. Delegacin de los Servicios Hidrulicos del Pirineo Oriental. Pla general dobres
per lany 1936. Annex a linforme. Arxiu Nacional de Catalunya. Fons 547. Agencia Catalana de lAigua. Unit 42.
48
CESALL Report. 1932. Arxiu Nacional de Catalunya (ANC) Fons 547. Agncia Catalana de l'Aigua. Unitat 479.
49
The letter from Provincial Governor of Barcelona was published in Appendix 2 of the CESALL Report. It should
be mentioned that the Chair of the CESALL Commission, and Director of the Water Resource Agency for the
Eastern Pyrenees, Enrique Gonzlez Granda y Silva abruptly resigned soon after this regulation was imposed on the
mining companies in September 1931. He was substituted on the CESALL commission by a representative from the
Mining District of Barcelona, Enrique Bayo Timmerhant, who chaired the remainder of investigation. A cynical
interpretation might see a connection between the unexpected water quality regulation, the immediate resignation of
the CESALL chair, and his replacement by someone from the Mining District of Barcelona, although no evidence
has been found to verify any connection.
50
CESALL Report.1932. Arxiu Nacional de Catalunya (ANC) Fons 547. Agncia Catalana de l'Aigua. Unitat 479.
43.
legislation that would limit chloride concentrations at 250 mg/L in the town of Pallej. This figure was

based on regulations inscribed in German law that oversaw the potash industry along the Rhine River.

However this apparent victory for Sociedad General Aguas de Barcelona, was in reality, a good deal for

the mining companies. Mining interests succeeded in ensuring that chloride concentrations were not

regulated at the point of discharge, nor even at the confluence of the two major rivers, as originally

established in the provisional Governors decree from 1931. Instead, the Commission pushed the

sampling site down river, nearly to the base of the watershed, where the dilution potential was greatest.

The sampling site of Pallej was justified because geologists estimated that the river recharged the aquifer

at this point (Fig 2.7).

44.
Figure 2.7 Map of the critical areas of the salinity conflict in the Llobregat watershed, published in the
weekly periodical El Mirador on June 27th 1935 in an expos article The problem of the salinity of the
waters of Barcelona. The hatched squares show the location of the potassium mines along the Llobregat
and Cardener Rivers in 1935: Unin Espaola de Explosivos, Minas de Potasa de Sria, Potsas
Ibricas, La Fondina y la Minera. The filtration zone where the Llobregat River feeds into the aquifer is
marked between the towns of Martorell and San Adrs de la Barca as Zona Filtrante. Finally, the
transparent square shows the location of the groundwater pumping wells owned by Sociedad General
Aguas de Barcelona in the town of Cornell, south-west of Barcelona.

Second, the CESALL commission recommended that water quality be monitored periodically

along the rivers. This was probably the most significant and tangible victory for those seeking to keep the

45.
aquifer salt free. Constant vigilance would be needed to enforce the new regulations. The monitoring

methods proposed were advanced for the era and included modern automatic measurement devices.

Third, the Commission advocated for the construction of a brine collector that would transport the

industrial effluent from the mines and pipe it to the Mediterranean. This proposal had the added

advantage that the new canal might also transport municipal and industrial wastewater, and thereby

protect Barcelonas drinking water from other contaminants.

Finally, the CESALL commission proposed the construction of dams along the Cardener and

Llobregat Rivers. These structures could help regulate river flows and dilute salt concentrations during

the dry season. Farmers could also benefit from the dams, and the CESALL commission was quick to

point out the benefits for agriculture and flood control.

The conclusions of the CESALL report fit the long term interests of both the mining and water

companies. Following the release of the CESALL report in April 1932, Sociedad General Aguas de

Barcelona became a firm advocate for the construction of the brine collector and the dam. Especially with

the dame proposal, the interests of the mining company overlapped with those of the water company.

Storing water in the mountains would provide a generous source of drinking water for the metropolitan

region, and releasing it during the dry season would dilute the contaminants and increase the volume of

water entering the groundwater wells managed by Sociedad General Aguas de Barcelona.

The CESALL commission was a laudable exercise in river basin planning and should be credited

with searching for constructive solutions to the conflict. The Spanish government sought to find

compatible uses among competing water users. With scientific detachment and quantitative precision, the

commission searched for practical ways to satisfy competing needs. What the CESALL did not foresee

however, was the delays that would prevent the brine collector from being built for several decades. To

the contrary, the CESALL commission was confident that the brine collector and dams would be built

soon. But the projects stalled, and during the remainder of 1932, and through 1934, the salinity

controversy largely disappeared from archival records and press reports. One major exception was the
46.
passage of the Salinity Law by the Catalan Generalitat published in August 1933. The law closely

followed the CESALL recommendations, essentially codifying their regulatory recommendations into

law.51 The regulation stated that the chloride concentrations could not surpass 250 mg/L along the

Llobregat River at Pallej.

Following the release of the CESALL report, the commission was renamed the Commission for

the Inspection of the Salinity of the Waters of the Llobregat (CISALL). As the new name implies, the

CISALL focused on monitoring water quality. With funding from the mining companies, the CISALL

established five sampling points, each below a mining operation on the Cardener or Llobregat Rivers: one

in Malagarriga, below the mines in Cardona; another in Antius, below Minas de Potasa de Sria; a third

in Torreroca, below the mines in Balsareny; a fourth at Soler Vicen below Potasas Ibricas; and the fifth

at the critical point of Pallej, the only site in which the data would have any legal implications.52

There is evidence that the Salinity Law and the regular inspections helped reduce the salinity

concentrations entering the freshwater aquifer downstream. In 1934 the water company reported that the

chloride concentrations fell from 202 mg/L to 169 mg/L, and they attributed this reduction to the Salinity

Law of 1933.53 However this success was short lived.

With the regulatory framework in place, attention shifted to the brine collector project which was

supposed to provide the definitive solution to the water pollution problem. However the project lost

momentum between the release of the CESALL report in 1932 and March 1935 when the issue finally

resurfaced. The exact reason for this delay is unclear. Leftist political parties governed during most of this

51
Butllet Oficial de la Generalitat de Catalunya. 1933. Llei de Salinitat. August 12, II, p. 229.
52
It appears that the Malagarriga station was not built until after the Civil War (Personal Comment, Santi Gorostiza,
October 2011)
53
Meeting Minutees CISALL, 25 April 1936 "La Sociedad que represento ve anulndose, poco a poco, los efectos
beneficiosos que en la disminucin de los cloruros contenidos en sus aguas se obtuvo con la inspeccin de los ros
Cardoner y Llobregat. Cuando empez dicha inspeccin, nuestras aguas de Cornell, contenan unos 202
miligramos por litro de in cloro; durante el primer ao y medio de inspeccin, disminuyeron hasta contener 169
miligramos por litro en Enero de 1935, pero desde dicha fecha, empez un ininterrumpido aumento (con tendencia
todava al alza, segn los ltimos anlisis), que ha hecho que la cloruracin se cifrara como promedio en el mes de
Marzo ltimo, en 187 miligramos por litro".
47.
period, both in Catalonia and Madrid, and one might expect Lefist parties to be more willing to take on a

polluting industry. Of course, this was a turbulent era, and many issues competed for the attention of civic

leaders. In the early 1930s north-eastern Spain was at the center of political turmoil and unrest. The

industrial areas of Catalonia were a hub of anarchist activity and occasional violent outbreaks.54 In

January of 1932, during a massive General Strike that spread throughout the Llobregat and Cardener

valleys, anarchists revolted in the mining towns and took control of the municipal government in Sria.

The anarchist flag flew over the city hall of Sria until the Spanish Civil Guards reclaimed control of the

town and expelled the rebels to African colonies.55 Following the revolt, the mining company saw the

need for more police protection and donated funds to the Spanish Civil Guard so they could establish a

permanent station near Manresa. Each mining enterprise in the region, Minas de Potasa de Sria, Unin

Espaola de Explosivos and Potasas Ibricas donated 12,500 Spanish pesetas to help build the police

quarters.56

The political turbulence certainly contributed to the limited attention directed at the salinity

problems in the Llobregat and Cardener Rivers, especially in 1934 when the issue nearly disappeared

from public view.57 However the water conflict resurfaced again in 1935, with considerable uproar, and

more attention from the press in Barcelona. In March 1935, the Director of Barcelonas Municipal

Laboratory, Dr. Pedro Gonzalez, released a report where he found that chloride values of Barcelonas

54
Brenan, G. 1943. The Spanish Labyrinth (12th Edition ed.). Cambridge: Cambridge University Press.
55
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
56
Arxiu Comarcal del Bages, Manresa.
57
I found surprisingly few documents that referred to the salinity conflict in 1934. In that year records show that the
water company Sociedad General Aguas de Barcelona opposed a request from Sociedad Potasas Ibricas to obtain
water concessions from the Llobregat River. On 19 August 1934 the La Vanguardia newspaper published a
photograph of Catalan President of the Generalitat, Lluis Companys, and the Catalan Comissioner of Public Health,
Josep Dencs, in an official visit to the mines operated by Sociedad Potasas Ibricas in Sallent. As mentioned,
there is evidence that effective water quality monitoring in 1934 helped reduce salinity values in the river and
aquifer. However in a letter dated 8 March 1935 from the water company executive Jos Soler Nolla to the
Governor of Catalonia, he asserts that the water quality monitoring is only helping to document the rate at which
chlorine is increasing in our rivers. He added that in January 1935 chlorine levels surpassed the permissible limit
and reached 272 mg/L in Pallej. The letter from Jos Soler Nolla can be found in Sallent Municipal Archives. Box
2525.
48.
drinking water had risen to 176 mg/L, meaning that concentrations had more than doubled in the last ten

years. 58 Dr. Gonzalez reiterated the recommendations from the CESALL report and was convinced that

the brine collector was the only solution.

The renewed attention directed at the salinity problems of the Llobregat River coincided with the

drought in the summer of 1935, in which reduced base flows increased mineral concentrations. 59 By

1935, records show that the mining companies were failing to comply with the Salinity Law of 1933.

Chloride concentrations at Pallej surpassed the limit of 250 mg/L and would often reach 300 mg/L.60

The deteriorating conditions of the Llobregat River also began to mobilize river users. In April 1935

fishers, writers and artists gathered in Manresa to protest the salinization of the Llobregat River. Fishers

from throughout Catalonia came to Manresa with their fishing poles in an event that the organizers called

Monster Meeting.61 Leaders from the anarcho-syndicalist movement spoke at the event and called for

the start of a water revolution.62

Unfortunately, little is known about the grassroots activism that protested the salinization of the

Cardener and Llobregats waters.63 But we do know that the conditions of the river had deteriorated

severely. Downstream users lamented that the river had become a sewer for industry. A resident of El

Prat del Llobregat lamented in 1935 that poets can no longer sing of their clear waters.64 Instead, the

58
La Vanguardia 1935. Dictamen del Director del Laboratorio Qumico Microbiolgico Municipal. 22 March 1935.
p 9.
59
Aldom, I. 2007. La lluita per l'aigua a Catalunya. De l'us, abs a la gesti integral (1900-2007). Lleida: Pags
Editors.
60
Letter from Jos Soller Nolla, Director of Sociedad General Aguas de Barcelona to the Mayor of Barcelona. 8
March 1935. Sallent Municipal Archive. Box 2525.
61
Miting Monstre Poster at Sallent Municipal Archive. Box 2525.
62
El Be Negre: Setmanari Satiric, 1935. La Revoluci de lAigua a Manresa. Num 200. 24 April 1935 p 2.
63
It appears that the President of the Federation of Fishers, Juan (Joan) Sebarroja was a strong defender of the
Llobregat River, and may have been a key grassroots organizer among the fisher community. His name appeared on
the poster that called for the protests in Manresa in April 1935. He also gave a lecture titled, The impurification of
the waters in Catalonias Rivers: Affair Llobregat in February 1935. (See La Vanguardia 21 February 1935 pag
13). In his lecture, he criticizes the poisoning of the river, and in particular, the extraordinary rise in salinity.
He expressed surprise that the authorities were not taking action to protect Barcelonas drinking water. To
demonstrate the economic contribution of the Llobregat River, he estimated that annually, fishers extracted fish from
the Llobregat worth 2 million Spanish Pesetas. In contrast, today, not a single edible fish is extracted.
64
Pujol, J. 1935. Agrcoles: Insistint. Noticiero Pratenc. 25 May 1935. Pg. 2.
49.
Llobregat had become turbid dirty, oily, milky, and even rotting during low flows.65 On the eve of

the 1929 World Exhibition in Barcelona, one newspaper worried that visitors might wander down to the

Llobregat River and find a pestilent valley that was an assault on the public health.66

The protesting fishers in Manresa had seen first-hand how river pollution was wiping away

aquatic life. While many fish kills probably went unreported, one made headlines in October of 1933.67

The floating dead fish covered the surface of the river, and the volume was so overwhelming that several

company towns were forced to halt production to pull out the dead fish from the turbines. In all, an

estimated 50 to 80 tons of dead fish floated down the river during the episode.68 While the first news

report published in Madrid initially attributed the fish kill to the potassium mines, another report

published in Barcelona later attributed it to storm water runoff from calcium mines.69

In 1935 it was clear that the mining activity had not reached its full potential and that the salinity

concentrations would continue to rise and threaten Barcelonas primary source of fresh water. 70 In the

spring of 1935 Sociedad General Aguas de Barcelona distributed water to one and a quarter million

residents. The water company wrote the Catalan Minister of Health and reminded the Catalan

Government of their undeniable mission: to uphold the law approved by the Catalan Parliament on 11

August 1933 that prohibits that the salinity of the Llobregat river at Pallej exceeds 250 mg/L. If the

problem were to remain unattended the outcome would be disturbing and tragic.71

The renewed pressure pushed the Spanish Government into action once again. In April 1935 the

Spanish Minister of Public Works, Rafael Guerra del Ro, called for a meeting in Madrid with top

65
Ibid.
66
LAven 1928, 12 February.
67
Heraldo de Madrid. 24 October 1933. Las aguas del Llobregat contaminadas. Al verter en ellas residuos de las
minas potsicas se ha producido un exceso de salinidad que determino la muerte enorme cantidad de peces. Pag 10.
68
Ibid.
69
La Vanguardia 1933. Por qu murieron los peces del Llobregat. El hidrato de cal. No hay peligro para los
habitantes de Barcelona. 2 November 1933. Pag 7.
70
La Vanguardia 1935. La Salinidad de las agaus del Llobregat. 20 March 1935 p. 5; La Vanguardia 1935.
Dictamen del Director del Laboratorio Qumico Microbiolgico Municipal. 22 March 1935 p 9; La Vanguardia
1935. La salinidad de las aguas del Llobregat. 23 March 1935 p 6.
71
AGBAR Archives. Box 5615 no. 36. Nota de la S.G.A.B. al Consejero Sanidad.
50.
executives from the water and mining companies72. They were accompanied by the members of the

CISALL Commission and the leadership of the Water Resource Agency of the Eastern Pyrenees. The

meeting marked a turning point for the salinity conflict because all the stakeholders finally agreed to act

upon the recommendations laid out by the CESALL Commission in 1932 including the building of the

brine collector. The CISALL commissioners estimated that the project would cost between 10 and 12

million Spanish Pesetas. Meeting minutes show that the CISALL Commission proposed that the expenses

be divided among the mining companies, the water company and the Ministry for Public Works.73 Jos

Soler Nolla from Sociedad General Aguas de Barcelona demonstrated his enthusiasm for the Ministers

proposal by offering to finance the project with private capital in order to accelerate construction. By the

end of the meeting, everyone agreed that the next step was to transform their ideas into budgets and

blueprints. Those in attendance regained confidence that the brine collector would be built; it was just a

matter of arranging the details.

The Public Minister for Public Works, Guerra del Ro framed the brine collector project within

the context of creating jobs for the working class. The Minister bragged in a press conference that he

hurled an ultimatum at the mining companies. Either they collaborated in the construction of a disposal

system for their industrial waste, or the Spanish Government would suspend the mining concessions and

bring all mining activity to a halt.74 This strong language was unprecedented and perhaps never repeated.

The timing of events suggests that key leaders within the Radical Party may be credited for

putting the brine collector project back on the agenda. Debate surrounding the salinity conflict hit the

press in Barcelona on a regular basis starting in March 1935, and remained a hot topic through the

72
Veu de Catalunya 1935. La Salinitat de les Aigues de Barcelona. 27 April 1935; Correo Catalan 1935. El
Ministerio de Obras Publicas da una detallada referencia del mismo, ampliada en sus aspectos tecnico por el
ingeniero D. Lorenzo Pardo. 27 April 1935. Num. 19229 p 31; Diario de Barcelona 1935. Delegacin de los
Servicios Hidraulicos del Pirineo Oriental. 28 April 1935. p 5; La Vanguardia 1935. Las Aguas del Llobregat. 5
May 1935 p 8;
73
Fbrega i Enfedaque, A. 2009. Cum Grano Salis. La Sal i la Potassa a Sria, 1185-1982. Iberpotash i Ajuntament
de Sria, pg 428.
74
Veu de Catalunya. 1935. La Salinitat de les Aigues de Barcelona. Manifestaciones del seor Guerra del Ro.
Saturday 27 April 1935.
51.
summer. This coincided with the Lerroux Government in Madrid, but also with the appointment or

election of Radical Party members in critical posts: Jaime Polo Otin at the Water Resources Agency of

the Eastern Pyrenees, Rafael Guerra del Ro in the Ministry of Public Works were both members of the

Radical Party, as was the Mayor of Barcelona, Joan Pich i Pon. The political alignment in Barcelona,

Catalonia and Madrid, may have contributed to the tangible steps toward the execution of the brine

collector project in the spring of 1935.75

Another interpretation might trace the renewed attention to the letters written by Sociedad

General Aguas de Barcelona on March 8th 1935 to the Governor of Catalonia, the Mayor of Barcelona,

and later on March 26th 1935 to the Spanish Parliament. These letters were signed by a coalition of

associations such as the Medical Academy of Barcelona, the Regional Federation of Fishers, and

agricultural unions.76 However the Sociedad General de Aguas de Barcelona had been pursuing this

project for years, so it is unclear what pushed the government into action in the spring of 1935, if not

something particular about those in power during that period.

After the Ministerial meeting in Madrid the salinity conflict continued to attract attention in the

press, and the executives at Minas de Potasa de Sria followed these developments closely.

Correspondence between the mine director, Norbert Fonthier and his supervisors in Brussels has left us

with insiders view of their deliberations. The mines monitored the Spanish press with special attention to

controversy or political resistance to their operations. News of the mines in the popular press was

immediately reported back to the headquarters in Belgium. In July of 1935 the Barcelona newspaper La

Publicitat released two political cartoons mocking the poor conditions of the Llobregat River and the

threat it posed to the citys water supply (Fig 2.8 and Fig 2.9). Fonthier despised improvisation, and

75
I searched correspondence from the office of the Mayor in Barcelona, Joan Pich i Pon, at the Municipal Archives
of Barcelona (Arxiu Municipal Contempornea de Barcelona) during 1935 but did not uncover anything related to
the salinity conflict. This, despite the reports in La Vanguardia that the Mayor had met with the Director of the
Municipal Laboratory Pedro Gonzalez in March 1935 to address the risk of salinization of Barcelonas water supply.
76
La Vanguardia 1935. A los seores Diputados Catalanes jefes de minoria en el Parlamento Espaol. 7 April 1935.
p 23.
52.
sought to be one step ahead of his rivals. When the Ministry for Economic Development sent a

government inspector without notice, Fonthier wrote to his supervisors baffled that such an inspection

would come with no warning.

In February of 1936 Sociedad General Aguas de Barcelona complained to the Governor of

Catalonia that the mining companies were continuing to dump their waste into the river and the

concentrations were violating maximum levels established in the Salinity Law of 1933. The water

company reported eight violations in December 1935 and six in January 1936. The mining companies

paid the associated fines but were otherwise unaffected.77 Sociedad General Aguas de Barcelona

requested that the government revoke their authorization of mining activity until the brine collector and

dam projects were completed. The water company submitted their letter to the Governor of Catalonia on

February 15th, 1936, the day before national elections in Spain. These would be the last elections of the

Second Spanish Republic. They handed a victory to the Popular Front, a coalition of Leftist parties.

However the victory would be short lived, as the Spanish military began to plot a revolt.

77
AGBAR Archives. Box 5615. Escrito al Gobernador General de Catalua Solicitando Pronta Construccion Canal
Colector. 15 Febrero 1936.
53.
Figure 2.8 The cartoon titled The waters of the Llobregat was published on the front page of the
Barcelona newspaper La Publicitat on July 9th 1935. The artist mocked the deteriorating water quality in
Barcelona by exclaiming: A Barcelona resident in 1938 Give me a soda, with potassium.

54.
Figure 2.9 A front page cartoon titled The waters of the Llobregat published in La Publicitat on July 12th
1935.
What does the Institute say about the salinity of our waters?
The Agrarian Institute is not dedicated to the humanities, and therefore we do not think anything,
because we do not need to think.

Spanish Civil War and Aftermath

The military General Francisco Franco fired the first shot of the Spanish Civil War on July 17th,

1936. The immediate objective of the coup dtat was to remove the Leftist parties from government

even though they had fairly won the election just months earlier. From the perspective of the Spanish

Right, only a military insurrection could restore order and unity to a disintegrating Spanish society.

According to this view, Spain was dangerously close to socialist revolution, and the democratic

experiment had gone too far in redistributing land and permitting anti-clerical violence.78 For the extreme

Left, the violent clash initiated by the military heralded the beginning of the Socialist Revolution. The

78
Brenan, G. 1943. The Spanish Labyrinth (12th Edition ed.) Cambridge: Cambridge University Press.
55.
moderate Republicans governing in Madrid were caught in the cross fire between radically divergent

ideologies from the Left and Right. With the political space for moderates shrinking, the Republicans

struggled to govern. And as Francos military troops advanced from southern Spain, the Republicans

most immediate priority was to salvage the democratic institutions of the Second Spanish Republic.

When the news of the military uprising reached the workers in Barcelona, revolutionaries of

every color sprung to arms. Anarcho-syndicalists, Marxists, Communists, Leninists, Trotskyists, and

Stalin loyalists took fighting positions behind barricades in the streets of Barcelona. George Orwell

described a celebratory mood in Barcelona as the war began in his book Homage to Catalonia. Soon the

revolutionaries began to exert considerable influence within the Catalan Generalitat, and the city was

decorated with canvas paintings of Marx, Lenin and Stalin that covered the faades of large buildings in

central squares. Businesses were collectivized by the workers, and Minas de Potasa de Sria and

Sociedad General Aguas de Barcelona were no exception. Norbert Fonthier was taken prisoner for

several days, and when released, he escaped with his family on a boat to Marseille, France along with

other blue-blooded Spaniards and foreigners.79 Sociedad General Aguas de Barcelona was collectivized

by the workers and renamed Aiges de Barcelona Empresa Collectivitzada.

Amid the revolution, the discussions about the brine collector continued. Comrade Carrerras,

Comrade Subirana and Comrade Capell replaced Fonthier as representatives of the mines in Sria, and

Comrades Tarafa, Batllori and Soucheiron replaced Soler Nolla at the water company. They gathered on

June 19th 1937 with the Catalan Generalitat to reaffirm their commitment to the waste disposal project.

The engineering plans for the collector had recently been completed on October 10th, 1936, a few months

into the Spanish Civil War. The project included drawings of the collector, measurements and route

demarcations which showed the path that the collector would travel between the mines and the

79
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
56.
Mediterranean.80 Depending on the geologic or topographic conditions, the engineers envisioned that the

collector would change its sectional profile. With the designs completed, the next step was to secure

funding so that construction could begin.

The communication between the mining companies and government officials in 1936 and 1937

show few hints of war. By February 1937 the Commission for the Inspection of the Salinity of the Waters

of the Llobregat (CISALL) had reached an agreement with each of the mining companies, old and new,

regarding their contribution to the collector project. Even late in the war, correspondence showed the

clear intent to build the collector soon.

The brine collector came very close to being built during the Spanish Republic. All that was

needed was more time. But Francos troops were faster. They marched towards Barcelona with the

military support from other Fascist governments in Europe. The democratic powers of the West remained

neutral in the conflict, and the defenders of the Spanish Republic were smothered.

The Franco Era

Barcelona fell to Francos troops in late January 1939, and the executives of the water and mining

companies returned to their posts while many members of the revolutionary government fled abroad,

mostly to France and Mexico. The Franco regime was quick to nullify any reforms associated with the

revolutionary governments that had briefly taken over during the war. The dictatorship issued a

resolution directed at the potassium industry whereby it declared all exports by the workers collectives to

have been illegal.81 However the collectivized mines were unable to maintain potassium production at

pre-war levels. Mining activity dropped precipitously between 1936 and 1939, and so did concentrations

80
I thank Santi Gorostiza for sharing images of this project with me that he obtained from the Archivo General de la
Administracin in Madrid.
81
BOE 1939. 17 March 1939. Nm 77. Pg. 1561
57.
of chloride in the Llobregat aquifer, from 183 mg/L to 110 mg/L.82 This correlation provided further

evidence linking the mining activity with water quality in the aquifer.83 In the eyes of many, including

Jos Soller Nolla, this was irrefutable proof that the rise in chloride concentrations observed in the aquifer

was caused by mining activity.

The brine collector project underwent a series of redesigns during the early Franco period. The

first project revision was published immediately, in 1939 The year of Victory as declared on the

projects cover.84 This design was approved by a Ministerial Order on August 28th, 1940, but the

complexity of the enterprise caused additional revisions that delayed it further.85 The project was

reformulated in 1943, and then again in 1953. Opposition to the project may have also contributed to its

delay. Agricultural interests at the base of the watershed were concerned that the brine collector would

consume water and subtract from the total volume coming downstream.86

Throughout the 1940s, chloride effluents from the mines continued to build up in the aquifer. As

mining activity accelerated, chloride concentrations in the aquifer started to climb and never looked back.

In 1945 they reached 190 mg/L surpassing pre-Civil War levels, and in October of 1949 they were nearly

300 mg/L. 87 Communications from the water company to government officials showed increasing signs

of frustration on the part of Sociedad General Aguas de Barcelona. After more than a decade of pleading,

82
AGBAR Laboratory Archive. Sociedad General Aguas de Barcelona. Escrito dirigido por esta Sociedad, en
Diciembre de 1949 al Excmo Sr. Alcalde de la Ciudad Relativo al Servicio de Abastacimiento de Agua, Actual y
Futuro y el Problema de la Salinidad de la Misma y Anejos al Escrito.
83
Gorostiza, S. 2010. El conflicto salino en el suministro de agua a Barcelona (1925 1940). Encuentro Cientfico
Salud y ciudades en Espaa, 1880-1940. Condiciones ambientales, niveles de vida e intervenciones sanitarias. 8 y 9
de julio de 2010, Barcelona.
84
Here I rely on documentation from the Archivo General de la Administracin that was generously shared with me
by Santiago Gorostiza.
85
A letter from AGBAR on 1 December 1949 to the Mayor of Barcelona cites that this project was approved in an
Ministerial Order on 27 July 1940, and that it was communicated on 5 August 1940. On 25 September 1940 the
project was made public, and on 16 March 1941 the Junta de Usuarios y Beneficiarios del Canal Colector was
created by the Ministerial Order of 14 April 1942.
86
Gorostiza, S. 2010. El conflicto salino en el suministro de agua a Barcelona (1925 1940). Encuentro Cientfico
Salud y ciudades en Espaa, 1880-1940. Condiciones ambientales, niveles de vida e intervenciones sanitarias. 8 y 9
de julio de 2010, Barcelona.
87
AGBAR Laboratory Archive. Sociedad General Aguas de Barcelona. Escrito dirigido por esta Sociedad, en
Diciembre de 1949 al Excmo Sr. Alcalde de la Ciudad Relativo al Servicio de Abastacimiento de Agua, Actual y
Futuro y el Problema de la Salinidad de la Misma y Anejos al Escrito.
58.
the Chair of the Board of Sociedad General Aguas de Barcelona warned the Mayor of Barcelona, if one

day our company provides the city of Barcelona with salt water, it will not be our fault nor due to our

negligence because we have done everything humanly possible to avoid this outcome since the

beginning.88

Fortunately for the mining companies, the pressure to build the collector was relieved somewhat

with the construction of the Sant Pon dam, built between 1949 and 1956. The dam regulated discharges

in the summer, and diluted salinity concentrations during months of low flow.

By 1950 Sociedad General Aguas de Barcelona supplied water for nearly 2 million residents.89

Water shortages in the summer of 1949 had caused water restrictions, and raised concerns about long

term water availability.90 Sociedad General Aguas de Barcelona began to explore additional water

sources and all options were on the table. In 1949 the water company asked the Mayor of Barcelona to

approach the Spanish Government with a proposal that would bring additional water resources to the city

from the neighboring Ter River.91 While this proposal was being considered, Sociedad General Aguas de

Barcelona needed to increase existing supplies in the short term in order to meet demand. Their solution

was to build a new water treatment plant that would withdraw surface water directly from the Llobregat

River at the base of the watershed in Sant Joan Desp. The new water treatment plant was inaugurated in

1955, and its opening marked a significant turning point for Barcelonas water supply. Until then,

Barcelona had relied exclusively on groundwater. The last time Barcelona residents drank directly from

surface water sources was during the Roman era, approximately 2000 years prior. The decision to collect

88
que si un da el agua de Barcelona llegase a beberse salada no sera ni por su culpa ni por negligencia de la
Sociedad, que haba hecho desde el nacimiento del problema cuanto humanamente caba hacer para impedirlo.
AGBAR Laboratory Archive. Sociedad General Aguas de Barcelona. Escrito dirigido por esta Sociedad, en
Diciembre de 1949 al Excmo Sr. Alcalde de la Ciudad Relativo al Servicio de Abastacimiento de Agua, Actual y
Futuro y el Problema de la Salinidad de la Misma y Anejos al Escrito.
89
AGBAR Laboratory Archive. Sociedad General Aguas de Barcelona. Escrito dirigido por esta Sociedad, en
Diciembre de 1949 al Excmo Sr. Alcalde de la Ciudad Relativo al Servicio de Abastacimiento de Agua, Actual y
Futuro y el Problema de la Salinidad de la Misma y Anejos al Escrito.
90
Ibid.
91
Ibid.
59.
the murky stream flow at the base of the watershed may be surprising, especially when one considers that

large cities usually seek out pure sources higher in the mountains. However selecting to locate the water

treatment plant at the base of the watershed was an unequivocal vote of confidence in favor of the new

water treatment technologies. Treatment engineers at Sociedad General Aguas de Barcelona were

absolutely certain that they could clarify and disinfect the water themselves, in the same way that is done

in the other large cities of the world, Paris, London, Amsterdam, etc92

A few practical considerations also help explain this decision. The planned site in Sant Joan

Desp was close to the groundwater pumping station in Cornell, making it convenient for personnel

working at the water company. Withdrawing surface water would also allow them to take advantage of

dam releases upstream.93 If the company only relied on groundwater sources, they would not be able to

capture dam releases very effectively. However the primary explanation for the unusual downstream

location of the water treatment facility probably had to do pre-existing water rights. By the 1950s most

water concessions had already been distributed to industrial and agricultural users upstream, and these

users aggressively defended their rights. This left the city of Barcelona to collect its drinking water from

the base of the watershed, even though its quality was questionable. Thus the strategic decision to build

the water treatment plant at the base of the watershed was a combination of technological hubris and

simply showing up late.

The decision to build the water treatment plant at the base of the watershed set in motion

additional measures to defend the water quality flowing down the Llobregat River, and this water quality

protection strategy relied heavily on engineered approaches. In the 1960s two major tributaries, the Anoia

92
AGBAR Laboratory Archive. Sociedad General Aguas de Barcelona. Escrito dirigido por esta Sociedad, en
Diciembre de 1949 al Excmo Sr. Alcalde de la Ciudad Relativo al Servicio de Abastacimiento de Agua, Actual y
Futuro y el Problema de la Salinidad de la Misma y Anejos al Escrito.
93
By this point the aquifer was noticing the impact of sea water intrusion. While the potassium mines had alleged
sea water intrusion as a source of salinity in the aquifer in the 1930s, an exhaustive revision by the CESALL
Commission discarded this possibility. A new source of surface water would allow the water company to reduce the
amount of groundwater pumping. This reduction in groundwater materialized and let led the chlorine concentrations
to drop slightly. See (Ferret 1985).
60.
and Rub, were diverted into pipes and channels that ran parallel to the Llobregat River until their

contents were released back into the main stem of the Llobregat River a few meters below the treatment

plant. This structural solution prevented both tributaries from releasing contaminats into the Llobregat

River. It also allowed nearby industries to dump waste into the tributaries without contaminating

Barcelonas drinking water supply. But this structural design had major limitations. The fabricated

channels could not contain the large volumes of water that flowed during storm events, so every time it

rained, polluted water inevitably overflowed into the Llobregat River and travelled towards the water

treatment facility. A river monitoring team called the River Police was dispatched by the water company

to visually monitor water quality on a regular basis. When they saw that the river carried an oil spill,

foam, or some other unusual contaminant, the River Police would radio downstream to warn managers of

the oncoming pollution and instruct them to shut down treatment. Storm events frequently caused water

treatment to shut down, as did the presence of detergents or heavy metals such as nickel, zinc, chrome or

copper from entering the facility. The halting of water treatment because of pollution problems was

common throughout the 1960s and remains frequent today.94

Given these water quality problems, it is not surprising that city managers eventually turned away

from the Llobregat to search for alternative and cleaner sources of freshwater. By 1964, the chloride

concentrations in the Llobregat reached 2,411 mg/L in Pallej, surpassing the maximum permissible

levels established in the Salinity Law of 1933 by a factor of ten.95 And in 1967 Barcelona began to draw

water from neighboring Ter watershed to the East. The new supply diminished the Llobregats relative

contribution to the citys total water portfolio. This further helped reduce pressure on the Spanish

Government and the mines to follow through on their commitment to build the brine collector.

94
Personal comment, Roger Lloret, August 2010.
95
Lloret, R. 2004. La qualitat de laigua del riu Llobregat. Un factor limitant del passat, un element clau per al
futur. In Prat, N and Tello, E. El Baix Llobregat: histria i actualitat ambiental dun riu. Centre destudis Comarcals
del Baix Llobregat. p. 92-141.
61.
In the 1970s, chemists discovered that the mining effluent also caused a public health problem. In

addition to emitting salts, the mines also released bromine, which when combined with organic matter

and chlorine, reacts to form a group of harmful disinfection byproducts called trihalomethanes. These

carcinogenic compounds would eventually be regulated by the European Union and Spanish drinking

water legislation, and their removal ultimately motivated the installation of the desalination systems.

Post Franco Democracy to Present

Stopping the salty effluent from flowing into the Cardener and Llobregat Rivers became a

political priority of the Catalan government once they gained regional autonomy from Madrid in the early

1980s. In 1982 the Catalan Generalitat approved the brine collector project, and in September 1988 the

collector project was finally completed, and began to divert salty effluent from the mines into a pipe that

transported the waste directly to base of the watershed.96 Water managers observed an immediate

reduction in chlorine concentrations in the river.97 And in a strange reversal of roles, the water company

Sociedad General Aguas de Barcelona became responsible for maintaining the brine collector. Even

though the water company manages the infrastructure, the collector still breaks frequently and releases

salt concentrate into the river. As recently as January 14th 2009, chloride concentrations in the river

reached 837 mg/L due to a break in the pipes, and water treatment managers in Sant Joan Desp were

forced to halt water treatment because of high salinity values. The growth of the mine tailings, now huge

mountains of salt, has also generated public discontent. Two mine tailings continue to grow in size. In the

96
The brine collector first dumped the brine back into the Llobregat River, but beyond the water treatment facility in
Sant Joan Desp. Not until a few years later, did they complete the final segment that took the brine to the
Mediterranean. The period in which highly concentrated brine ran through the final segment of the Llobregat,
between the water treatment facility and the Mediterranean, may have also contributed to salt infiltration into the
aquifer (Personal Comment, Roger Lloret, 2011).
Also see: http://www.lasequia.org/montsalat/Impactes/Salinitzacio.htm
97
Estevan, A. and N. Prat. 2006. Alternativas para la gestin del agua en Catalua: Una visin desde la perspectiva
de la nueva cultura del agua. Fundacin Nueva Cultura del Agua. Zaragoza and Bakeaz: Bilbao, Spain.
62.
town of Sallent, residents have organized opposition to the continued growth of the salt mountains and the

contamination problems they cause.98

At its core, the salinity conflict in the Llobregat watershed was a struggle to control the rivers

services. For the owners of industry, the Cardener and Llobregat rivers generated valuable electricity that

allowed them to run their textile machines. For farmers the rivers waters increased crop production. The

water company relied on the Llobregat to replenish the aquifer and maintain a steady supply of drinking

water for its customers. And finally, for the mining companies, the rivers provided a free disposal system

that carried away their waste to the Mediterranean. These ecosystem services: food and water production,

energy generation, and waste disposal, came into conflict, and each user maneuvered to protect the

services that served their interests.

The salinity conflict in the Llobregat watershed is a critical piece of history that helps explain

why Barcelona now relies on two desalination facilities to meet its drinking water needs. The solutions

put forward to address this conflict demonstrated confidence in engineered methods to meet their water

challenges, predominantly in the form of dams and pipes. This view dominated the management of water

resources in Spain for much of the twentieth century. However we find an exception to this vision in the

writings of Rafael Puig Valls and his proposal for addressing flooding problems with reforestation and

ecosystem restoration. His proposal to manage ecosystems stood in opposition to the mainstream

structural solutions, and may be considered as a precursor to contemporary advocates of ecosystem

services.

98
See www.prousal.org
63.
Figure 2.10 It is estimated that this photograph was taken between 1932-1936. It shows the small dam
along the Cardener River that diverted water for the Antius factory owned by Antonia Burs Borrs and
Esteve Torrens Burs. Antius is the first textile factory below the potash mines, located less than a mile
upstream. The small structure in the lower left corner no longer exists. However it is possible, although
not confirmed, that this structure housed the automatic water quality instruments established by the
CISALL to monitor salinity values in the Cardener. (Photograph courtesy of Albert Fbregas)

64.
Figure 2.11 In the 1960s employees of Sociedad General Aguas de Barcelona regularly travelled
upstream to check on the instruments that measured river water quality. The names of the sampling sites
established by the CISALL commission remain the same, even if the exact locations have changed
slightly. This picture taken in the 1960s shows a visit by AGBAR employees to the water quality
measurement station located on the Antius canal next to the Cardener River. The original automatic
station, possibly depicted in Fig 2.10, might have been washed away by a flood, and then moved to its
current location midway along the canal. (Photograph courtesy of Roger Lloret)

Figure 2.12 The Antius water quality station in April 2011. The Catalan Water Agency still uses the
instruments in this structure to collect water quality measurements. Today, the vegetation is so high that
neither the canal nor the river can be seen from this point. Note the rapid growth of the pine tree next to
the installation. (Photograph by the author)

65.
2.4 Ecosystem Services for Flood Management
For centuries, residents living near the Llobregat River feared the destruction caused by the

rivers floods. Apprehension and anxiety would probably best describe their relationship with the

Llobregat, a flumen terrible et periculosum.99 This apprehension was diluted only by a firm desire to tame

its waters and protect their property. The constant threat of flooding conditioned development around the

river. Overflowing waters threatened to wash away the winters food supply, essential bridges that

connected goods to market, homes, barns or livestock. Residents in the town of Sant Boi de Llobregat

pioneered flood control in 1597 by building a stone wall along the banks of the Llobregat River.

However these walls did not last long, as the river flooded them out repeatedly. A few years later, in the

early 1600s, the towns in the lower Llobregat joined forces to build a communal levee system. Alas, they

built the levees too close to the river, suffocating its flood plain, and this system too failed to withstand

the rivers strength. In 1638 the collaborative venture started anew, and this time they conceded the river

a 200 meter flood plain on each side in an effort that, records show, required considerable financial

sacrifice on the part of each collaborating municipality. But these structures too failed to contain the

Llobregat, and the floods continued.100

By the late 1800s, engineers began to look upstream for flood control measures. They surmised

that if they could build a dam in the mid section of the watershed it would allow them to retain peak flows

and prevent flood damage. The waters held behind the dam could be released gradually when the risk of

flooding had past. A dam-oriented solution to the Llobregats flooding problems would also permit the

administration to store water for the dry season. Farmers were especially interested in water for irrigation

during the summer months, and the owners of the hydro-powered textile mills sought reliable flows to

maintain a steady energy supply.

99
Terrible and dangerous river Codina, J. 1971b. Inundacions al delta del Llobregat. Episodis de la histria. Vol.
147-148. Barcelona: Rafael Dalmau.p 74.
100
Codina, J. 1971. El delta del Llobregat i Barcelona. Gneres i formes de vida dels segles XVI al XX. Hores de
Catalunya. Esplugues de Llobregat: Edicions Ariel.
66.
By 1890, an engineer named Salvador Peydr had drawn up the plans to construct a dam along

the main stem of the Llobregat near Cabrianes. The project was designed to flood 400 hectares and retain

48 million cubic meters of water. The goal was to prevent flood damage by regulating flows, while at the

same time increase irrigation supplies. Proponents of the dam also argued that it could provide the city of

Barcelona with an additional source of drinking water. Peydr estimated that the dam would supply the

lower Llobregat with a steady flow of 8 cubic meters per second year round. This would be a huge

improvement, since between the months of June and September the discharge in the Llobregat often

dipped below 4 cubic meters. When the project was submitted, it was Spains largest hydraulic venture

on record, with a total cost of 28 million Spanish Pesetas.101

The engineered approach to flood management found its opposition from a forester, Rafael Puig

Valls, an avid field scientist and naturalist. Later in his career, in 1904, Puig Valls would become the first

to advocate for the creation of protected areas in Catalonia: the mountains of Montserrat, the Montseny

and the Collserola ridge that hovers over Barcelona.102 But during the debate concerning the Cabrianes

dam, Puig Valls merely spoke as a forester who understood the relationship between forest ecosystems

and hydrologic flows. His criticism of the dam project emerged from his acute observations of the

Llobregat River throughout the seasons. He saw that the Llobregat had been transformed by deforestation

upstream and hypothesized that the flow regime observed in 1890 had already been modified by human

activities. For instance, he felt that the severe channel incision in the Llobregat was the result of higher

flows concentrated in shorter periods. Salvador Peydr and Rafael Puig Valls observed the same flooding

events in the Llobregat watershed, and yet they drew remarkably different conclusions. Whereas Peydr

characterized the torrential floods as part of a natural flow regime, Puig Valls concluded that the floods

were the artifice of poor forest management in the Pyrenees Mountains. Puig Valls recognized that

101
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona.
102
Boada, M. 1995. Rafael Puig i Valls (1845-1920) Precurso de l'educaci ambiental i dels espais naturals
protegits. Barcelona: Departament de Medi Ambient, Generalitat de Catalunya. Born in 1845, Puig Valls studied
forestry. Following a trip to the United States, he returned to promote the celebration of Arbor Day.
67.
Mediterranean streams naturally had erratic discharge, in fact, he emphasized this irregularity as a central

feature of Mediterranean streams. But Puig Valls also stressed that recent flooding events had become

more severe as a result of human modifications to the watershed and disruptions to the hydrological cycle.

In several essays published by the Royal Academy of Sciences and Arts of Barcelona, Puig Valls

applied his ecological and hydrological expertise to criticize the Cabrianes dam project. For Puig Valls,

the dam proposal was excessively ambitious, expensive and unnecessary. For the same financial

investment, he argued, residents could achieve better results by investing in the restoration of forest

ecosystems in the upper segment of the watershed, complimented with only smaller flood control

structures. In contrast to the proposal offered by Salvador Peydr, Puig Valls argued that reforestation

and ecological restoration could regulate the waters flow without relying on the building of dams or on

the specialized procedures of the domain of the construction industry.103 This positioned Puig Valls as

the first in Spain to advocate for flood control measures that adopted an ecosystem approach.

When Puig Valls surveyed the upper Llobregat watershed, he lamented that severe deforestation

and grazing were causing massive erosion and flooding. The water precipitates onto the steep slopes of

the upper Llobregat watershed, and when it falls on the denuded land and exposed rocks, the waters fall

with a thunder, and rapidly searches for the channel toward the sea. 104 This trend needed to be reversed.

We must research then, how to modify the effects of evaporation and retain these waters in its course in

order to transform the rapid and violent flushes of water into constant flows, slow and calm, that is to say,

103
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pg 36
104
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pg 379
Caen las aguas en las Fuertes pendientes de la Cuenca alta del Llobregat y al encontrar el suelo desarbolado, y la
roca al descubierto, las aguas se precipitan con estrpito y corren veloces a buscar la vaguada principal hasta
llegar al mar.
68.
we must normalize the flow regime in our waterways, both in the main stem and its tributaries of the

Llobregat which today has only a torrential flow.105

Puig Valls praised forests for regulating the Llobregats waters. His essays went into great detail

to describe the hydrologic cycle in a forest ecosystem. The forest slowed and retained water, cushioning

the rain drops fall on its leaves, travelling along its branches and trunks, slowed again by the forest

understory, and then filtering into the soil and aquifer. He contrasted waters leisurely travel through the

forest with the speed and strength that waters would develop in denuded areas, where regardless of how

gentle the slope, the runoff would gush rapidly into the streams and swell them quickly. He compared

the retention capacity of forests to a natural reservoir or sponge that gradually released water throughout

the seasons. Forested areas withhold large amounts of water that is retained by the roots of the

vegetation, which, having acted like a sponge, do not let a single drop run along the surface, and do not

create a single ditch.106 He also pointed out that the forests shaded the earth, protecting the soil from the

hot sun, and maintained humidity beneath the vegetation. Furthermore, forests helped maintain a steady

flow in mountain springs since the slow percolation of water through the leaves, organic matter, and soil

permitted a regular flow throughout the season.

Puig Valls was not the first to observe the relationship between forests and water flow, and in his

writings we see considerable influences from George Perkins Marsh. Sixteen years prior to the

publication of Puig Valls essay on the Llobregat, Marsh had written The Earth as Modified by Human

105
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pg 379
Hemos de investigar, pues, la manera de modificar los efectos de la evaporacin, y retener las aguas en su curso
para convertir los desagues rpidos y violentos en corrientes constantes, lentas y tranquilas, es decir, hemos de
normalizar el rgimen de las aguas en los cauces del ro principal y de los afluentes del Llobregat que tienen hoy
carcter torrencial.
106
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pg 380
La superficie arbolada presenta todos los caracters de un suelo abrigado, de evaporacin lenta que retiene en su
masa una gran cantidad de agua que absorben las races de las plantas, y que habiendo hecho oficio de esponja, no
han dejado correr una sola gota de ella por la superficie, que no presenta como surco alguno
69.
Actions while serving as an American diplomat in Rome.107 As a diplomat and linguist who specialized

in the Icelandic language, Marsh was an unlikely figure to write the first treatise on environmental

protection. But his work caught the attention of scholars and intellectuals, and today he is credited with

starting the conservationist movement in the United States. Puig Valls must have read the writings by

Marsh and been influenced by his work.

When Marsh was writing in the late nineteenth century, pioneers in the United States were

moving west, to the arid states of Colorado, Arizona and California. But while his contemporaries

travelled west, Marsh went east, to the aging European continent, where he would read and write about

the past. Stationed in Rome, surrounded by the eroded columns of an extinguished civilization, Marsh

was intrigued by the causes of Romes collapse. His linguistic dexterity allowed him to devour texts in

Latin, Italian, French, German, and Spanish; all of which uncovered a series of environmental fiascos,

which in some way contributed to the decline of their civilization.

Marsh recounted examples in which deforestation led to diminished stream flow or perishing

springs. Puig Valls took this one step further by applying this knowledge to oppose a dam project and

advocate for ecological restoration. The reforestation proposal became his alternative to the structural

approaches offered by Peydr. From our historical distance, we can now appreciate that Puig Valls was

advocating for the management of forests for ecosystem services. Furthermore, Puig Valls claimed that

the management of the watershed would allow the Government to meet the same objectives at a lower

cost. Puig Valls faith in ecosystems contrasted with his skepticism of engineered solutions. He warned

that the rectification work of the Llobregat River in Molins de Rei would be deficient, incomplete and

useless if they were not supported by improvements in the upper segment of the watershed.108 Over

107
Marsh was exceedingly educated and well read. Born in Vermont in 1801, he studied at Dartmouth, and learned
to speak many languages. In 1849, President Zachery Taylor sent him on a diplomatic mission to represent the
United States before the Ottoman Empire. Later Marsh was sent to Greece, and in 1861 President Abraham Lincoln
dispatched him to Rome as the American Ambassador in Italy.
108
Puig i Valls, R. 1904. El Llobregat: sus cuencas alta media y baja y obras indispensables que hay que realizar en
ellas para conseguir que las inundaciones sean cada vez menos temibles y las aguas normales ms constantes con
70.
time, Puig Valls became more explicit. In 1904 he wrote, it is evident that if forests occupied large parts

of the upper Llobregat watershed the effects from flooding would be reduced because of the forests

powerful influence as an agent that retains, filters, and captures the waters that fall on its slopes.109

Eventually, the Cabrianes dam project was discarded because of resistance from the flooded

municipalities. The opposition from Puig Valls was probably a footnote in the final decision to table the

project, however as we will see, his ideas remained influential in discussions on flood management in the

Pyrenees Mountains.

Around the time of Puig Valls writings in 1890, land managers in the United States had also

noticed that forest ecosystems helped retain hydrologic flows. One of the early advocates for hydro-

reforestation works in California was William Mulholland, the Public Water Commissioner of Los

Angeles at the turn of the century. Mulhollands leadership of the hydraulic projects that brought water to

the city of Los Angeles has been immortalized by Marc Reisners book Cadillac Desert, and the movie

Chinatown. Therefore it may be surprising to learn that Mulholland initially opposed dams, and surface

water transport systems in general, because of the water lost to evaporation. For Mulholland groundwater

was more reliable and efficient. This led him to advocate for the reforestation of mountains in California

to improve water retention and infiltration. And when the former Mayor of Los Angeles approached

Mulholland with the idea of transporting water from the Owens Valley on the Eastern face of the Sierra

aumentos de riqueza pblica y particular. Memorias de la Real Academia de Ciencias y Artes de Barcelona, tecera
poca. Vol. IV, nm. 40. Barcelona. p 536.
109
Puig i Valls, R. 1904. El Llobregat: sus cuencas alta, media y baja y obras indispensables que hay que realizar en
ellas, para conseguir que las inundaciones sean cada vez menos temibles, y las aguas normales ms constantes, con
aumentos de riqueza pblica y particular. Memorias de la Real Academia de Ciencias y Artes de Barcelona, tecera
poca. Vol. IV, nm. 40. Barcelona. p 427
Claro es y evidente que, si existieran grandes superficies arbolados en la Cuenca alta del Llobregat, los efectos de
la inundacin seran menores, porque las influencias poderosas de su accin, como agentes que retienen, filtran y
detienen las aguas en su cada sobre las vertientes
71.
Nevadas to Los Angeles, Mulholland originally thought the idea absurd.110 Only later, the ex-Mayor of

Los Angeles convinced Mulholland of the scheme, and he became the projects lead promoter.

The United States Senate also recognized the contribution of forests in stabilizing water flows. In

the Organic Administration Act of 1897, the same Act that created the US Forest Service, it states that

forests could be protected for the purpose of creating favorable conditions of water flow.111 Thus

forests contribution to water management was appreciated in Europe as well as the United States.

In 1907 a tremendous flood ripped through the Llobregat basin, washing away many industries in

the Cardener and Llobregat valleys. This disaster re-energized concerns about flooding in the Llobregat

valley. Motivated by this recent flood, the Spanish Cortes asked Senator Duran Ventosa to chair an

investigation into the Llobregats flooding problems. His recommendations published in 1914 picked up

on ideas put forward by Puig Valls to mitigate flooding through reforestation.112

Duran Ventosa emphasized that towns and villages depended on the Llobregat for their well-

being. He spoke of collaboration with the river, and of developing activities that were appropriate with

the rivers needs. Like a sick patient, it was necessary to dedicate the necessary attention to the

Llobregats problems, but not only during times of floods, but also under regular conditions, when

problems can be resolved with more sobriety and veritable efficacy.113 The Senators report came

twenty four years after Puig Valls first proposed hydro-reforestation projects for the Llobregat basin and

advocated for an aggressive reforestation program. But the Catalan Government was not able to act on

these proposals until the return of the Spanish Republic in the 1930s.114 Starting in 1933 the Catalan

110
Reisner, M. 1993. Cadillac Desert: The American West and its disappearing water. Penguin Books: New York. p
61
111
Ibid. p 83
112
Duran Ventosa, L. 1914. Memoria sobre alguns problemes ques presenten en el riu Llobregat. Barcelona:
Diputaci Provincial de Barcelona. Biblioteca de Catalunya.
113
Ibid.
114
In 1914 the Catalan government still had limited autonomy under the Spanish Crown.
72.
Generalitat began to fund reforestation projects in the Pyrenees with the explicit goal of regulating water

flows in the Llobregat Basin.115

The debate over flood management in the Llobregat watershed brings into sharp relief the

competing approaches to river management. Puig Valls and Duran Ventosa interpreted the floods as

symptomatic of poor ecosystem management whereas dam advocates asserted that the new structural

designs would allow them to build a definitive solution to flooding problems. The competing camps also

relied on different schools of knowledge to support their position; some favored knowledge about

structures and cement, while others valued knowledge about forests and soils.

The writings of George Perkins Marsh, Rafael Puig Valls and Duran Ventosa show us that ideas

about ecosystem services are hardly new. These ideas were even put into practice by the Spanish

Republic in the 1930s. But ultimately, approaches that favored ecosystem management were pushed

aside for engineering strategies that sought to dominate water resources through structural design.

It is difficult to pin-point exactly why notions of ecosystem services were pushed aside in favor of

structural designs. Perhaps it was because projects relying on ecosystem management did not pull

together a coalition of political and economic interests to support its execution. In contrast, structural

solutions usually had strong backers who would benefit greatly from the contract and execution. Puig

Valls even refers to these economic interests in his writings. And globally, dam builders have yielded

considerable political influence.116 I have not undertaken a detailed investigated into the political

relationships between dam builders and political powers in the Llobregat watershed at the beginning of

the twentieth century. But I did uncover at least one instance in where there was a connection. Joan Pich i

Pon, the Mayor of Barcelona (1935), and later Civil Governor of Catalonia, built his financial fortune

with the construction of dams in the Pyrenees and selling the electricity they produced. His company

Electricadad del Cad won several contracts, after which, he decided to enter politics. In 1929, prior to

115
(Boletin Oficial de la Provincia de Barcelona 1933)
116
McCully, P. 2001. Silenced Rivers. The ecology and politics of large dams. Zed Books, New York.
73.
becoming Mayor of Barcelona, Pich i Pon secured water rights to build a hydroelectric dam along the

Cardener River.117 His project never came to fruition, even though the dam was eventually built by

someone else 30 years later. Therefore I suspect that there must have been more connections between the

political and economic elites that would favor the building of large infrastructure projects. The mining

companies took care to hire advisors in Madrid to stay informed with political and regulatory

developments, and the same was probably true for proposals surrounding water management. Without the

confluence of interests, it was much more difficult for projects to be developed.

Figure 2.13 Rafael Puig Valls (1845-1920). (Image from Boada 1995)

117
Gaceta de Madrid, 1929. Ministerio de Fomento. Direccin General de Obras Pblicas. 24 Febrero. Num 55. p.
1479.
74.
2.5 Conclusion
Nature tells us that we must look at the origin of our ills to find our cure.118

Rafael Puig Valls - 1904

Reviewing the struggle over salinity values in the Llobregat River has helped us identify the

major figures and institutions in the watershed: the industrialists that harnessed the rivers power; the

private water company Sociedad General Aguas de Barcelona; the Water Resources Agency of the

Eastern Pyrenees; Spanish Ministry of Public Works; the Regional Government of Catalonia; and the

mining companies.

The conflict also illustrated how these stakeholders made decisions about their water resources.

The degree of democratic participation in these decisions evolved in step with the political changes taking

place in Spain. During the Primo de Rivera Dictatorship, power was centralized and the pollution

problems of the Cardener and Llobregat Rivers were delegated to a small group of experts with close ties

to the mining industry. A new formula did not appear until Spain transitioned to a Republican

democracy. The CESALL Commission began to operate during a period of democratic governance, and

they soon invited the competing parties to participate in the search for solutions.

It should be recognized that the CESALL Commission was a remarkable institution for its time,

and they undertook the first watershed planning process in the Llobregat basin. This planning process

revealed the competing visions for the rivers future as rival water users sought to secure the river-related

services that most benefited them.119 But the commission members also sought to resolve these conflicts

with the best science available. The commission meticulously collected water quality samples, they

reviewed the new data, and consulted experts from Germany and abroad.

118
Y que en el origen del mal cabe buscar remedio, lo dice la misma naturaleza.
119
Not all stakeholders were represented in the CESALL commission. Fishing interests were notably absent. But
the CESALL commission at least began with a public request for information so the public could inform the
deliberations.
75.
It is difficult to argue the salinity conflict was not resolved because of an inappropriate

institutional framework for watershed management. Instead I find that it was precisely the type of

consensus that emerged from the CESALL commission that set Barcelona on a path of hydraulic

dominance. Agreement was reached on various fronts, but first and foremost, the brine collector project

emerged as the definitive solution to the salinity problems of the Llobregat River. The regulatory

approach was only an emergency measure; designed to hold back chloride concentrations until the

structural solutions were put in place. The Salinity Law of 1933 was loosely enforced as chloride

concentrations regularly surpassed the maximum value of 250 mg/L starting in 1935. The lax

enforcement suggests that the regulatory solution was never taken seriously. Instead of managing the

pollutants at their source, governing officials in Barcelona and Madrid placed their bets on the brine

collector, which together with the dams, would resolve the salinity problems in the Llobregat River.

The outcome of this planning process helps us understand how Barcelona was sent down a path

of expensive and energy intensive water treatment. More often than not, major decisions in the watershed

were guided by a faith in engineered solutions. This is evident in the decision to build the brine collector;

raise the dams; withdraw water at the base of the watershed, and most recently, install desalination

systems. All of these decisions share the conviction that technological mastery would solve Barcelonas

water problems. These choices are in fact part of a continuum; evolving in step with technological

changes, but yet repetitive because of a shared underlying logic.

The historian Donald Worster has described the irrigation society of the American West as

trapped by their own inventions, encased in its past and unable to escape its own history.120

Barcelonas hydraulic system is no different. Water managers have inherited dams, canals, pipes and

membranes that pushes them further along a trajectory already chosen by their predecessors. They also

have inherited certain notions about how the river should be managed. Over time, these beliefs go

unquestioned, and they create new material realities.


120
Worster, D. 1985. Rivers of Empire. Cambridge University Press. London. p 329.
76.
Visit the desalination system in Sant Joan Desp and you can touch the colorful pipes feeding the

reverse osmosis membranes. At the drinking water plant in Abrera you can see the batteries that send

electrical pulses through the water to pull out charged ions from the rivers water. These desalination

systems are the logical continuation of a belief system based on technical fixes and hydraulic dominance.

The energy intensive systems form part of a long tradition of technological remedies that attempt to solve

Barcelonas water treatment challenges with the latest gizmos.

Contemporary praise for reverse osmosis desalination as the ultimate barrier in water treatment

is reminiscent of the reverence bestowed on the brine collector project as the definitive solution. And

this conviction that technology will resolve the problems of the Llobregat River has diverted attention

away from the upstream sources of the predicament. Even when the brine collector was completed in

1988, fifty six years after it was proposed by the CESALL commission, water managers remained

burdened by a cocktail of contaminants that arrive at the entry point of the two major drinking water

facilities. Thus the technological improvements have arguably permitted the mines to continue their

polluting activities without addressing the issue at the source.

The history of the salt pollution also informs a debate about the origins of the salts found in the

Cardener and Llobregat Rivers. In 1932 not a single member of the CESALL commission denied that the

potassium mines were dumping salty effluent into the river. If anything, the questions pertained to how

much was being released and if those effluents could be causally linked with the rise in salinity in the

aquifer down below. This history provides unambiguous evidence that the salinity in the Llobregats

waters is the product of human activities, and yet this conclusion would be redundantly obvious to water

managers in the 1930s.

Without a clear historical account of a rivers pollution problems, it is easy for competing

narratives fill the void. Mining interests today insist that the chloride found in the Llobregat River comes

from natural sources. And this narrative has gained traction among reputable institutions. In a document

written by a regional planning agency in the 1980s, the Llobregat River is described to have high
77.
concentrations of salts due to the dissolution of salt rocks that are mined in the watershed.121 In other

words, the high chloride values can be attributed to the protruding salt diapers in Cardona. The report

goes on to explain that the Llobregat aquifer at the time (1980s) was extracting water with chloride

concentrations of 600 mg/L, far above the 79 mg/L detected in 1925, or the legal limit in Palleja of 250

mg/L. Nevertheless, these chloride concentrations were deemed acceptable.122 It must be recognized

that some portion of the chloride content in the rivers can be attributable to the natural geology. But it is

equally undeniable that mining activity has caused the severe salinization of the Llobregat River and its

aquifer.

Efforts to dominate rivers have overshadowed attempts to seek more symbiotic relationships with

river ecosystems. Puig Valls appreciated how forest ecosystems could help retain water in the upper parts

of the watershed through ecosystem processes that had no need for human intervention, only the

restoration of ecosystem functions that had been lost. This makes contemporary advocates of ecosystem

services heirs to the ideals proposed by Rafael Puig Valls and George Perkins Marsh. They sought to

reawaken society to our dependence on ecosystem processes. Similarly, the field of ecosystem services

invites us to reexamine our relationship with natural systems and broaden the management options

available for addressing sustainability challenges.

So even if ideas about ecosystem services are not entirely new, they still diverge from mainstream

approaches for water treatment and distribution. Few are proposing an ecosystem services approach to

address the watershed problems found in the Llobregat today. And the idea that ecosystem processes can

be of value to downstream users has often been met with skepticism by local users. Perhaps this is

because ecosystem-based approaches to water management run contrary to the training provided in the

121
ro Llobregat, ro que como se sabe lleva aguas con alto contenido de sales debido a la disolucin de rocas
salferas de la Cuenca minera de la depresin central catalana. Corporacin Metropolitana de Barcelona. 1986.
Usos agrcolas de los mrgenes y delta del Llobregat. Barcelona. p 21.
122
Corporacin Metropolitana de Barcelona. 1986. Usos agrcolas de los mrgenes y delta del Llobregat. Barcelona.
p 29.

78.
water treatment field. Aspiring water treatment managers are taught to master mass-balance equations, or

dominate techniques for coagulation, flocculation, filtration, disinfection or desalination. Understanding

many ecosystem processes in freshwater rivers remains outside their field. In the field of water treatment,

notions about ecosystem services might offer a useful and interdisciplinary perspective to address water

treatment problems.

Finally, historical analysis offers us an opportunity for transformation. Historical examination is

especially useful to interrogate mainstream notions that go unquestioned. Without a firm historical

foundation, we may be tempted to believe that current dysfunctions are the natural state of affairs rather

than an exception. We might also submit to the status quo, and accept that our rivers have always been

managed this way. At worst, we might surrender to the poor state of our rivers instead of working for a

cleaner future.

79.
2.6 Acknowledgements
Many have helped me develop this historical research. Good friends and strangers have pointed me in the

right direction and selflessly answered my questions. In particular, I would like to thank my advisor,

Daniel Schneider for reading several drafts and providing suggestions for improvement; Santiago

Gorostiza for sharing notes and documentation from the AGBAR Archives as well as providing valuable

feedback on drafts; Roger Lloret for his vivid stories about key events in the watershed and for

connecting me with other knowledgeable individuals, Esteve Torrens Prez de los Cabos, grandson of

Antnia Burs Borrs, for the detailed description of his grandmother and for family photos; Albert

Fbrega Enfedaque for selflessly sharing his insight and information about the history of mining in Sria;

Antonio Snchez, for his gift of the book Cum Grano Salis and for his willingness to answer my

questions; Sara Ortiz Escalante, for help in translations; Jordi Mir for sharing his insight about Spanish

trains of the era; Enric Toms Guix for providing tips on how to conduct historical research; the

Municipal Archives of Sallent and Sria for allowing me to reproduce material; the Arxiu Comarcal in

Vilafranca del Peneds; and the librarians at the Biblioteca de Catalunya and Arxiu Nacional de

Catalunya for helping me find relevant primary sources.

80.
2.7 References Chapter 2

Alabern, et al. 1991. Historia de la Ciutat de Manresa (1900-1950). Caixa de Manresa. Manresa, Spain.
Aldom, I. 2007. La lluita per l'aigua a Catalunya. De l'us, abs a la gesti integral (1900-2007). Pags Editors.
Lleida, Spain.
Arnau Reitg, R. 1981. La mineria del Bages: una visi retrospectiva. XXVI Assemblea Intercomarcal d'estudiosos a
Manresa (pp. 53-58). Centre d'Estudis del Bages. Manresa, Spain.
Balcells, Albert. 1998. De la crisi del rgim autonmic a laixecament military del 1936. In Histria de Catalunya.
Volum XI. 1586-1600. Salvat Editores: Barcelona, Spain.
Boada, M. 1995. Rafael Puig i Valls (1845-1920) Precurso de l'educaci ambiental i dels espais naturals protegits.
Departament de Medi Ambient, Generalitat de Catalunya. Barcelona, Spain.
Bolaos, A. 2004. Aiges de Barcelona, histria dels reptes per al subministrament daigua. In 214-225. In Prat, N
and Tello, E. El Baix Llobregat: histria i actualitat ambiental dun riu. Centre destudis Comarcals del Baix
Llobregat. Barcelona, Spain.
Brenan, G. 1943. The Spanish Labyrinth (12th Edition ed.) Cambridge University Press: Cambridge.
Carol, M. 1991. El Llobregat: Un Cam daigua. Langwerg Editions. Diputaci de Barcelona. Barcelona, Spain.
Cioc, M. 2002. The Rhine. An Eco-biography, 1815-200. University of Washington Press: Seattle, WA.
Corporacin Metropolitana de Barcelona. 1986. Usos agrcolas de los mrgenes y delta del Llobregat. Barcelona.
Cronon, W. 1995. Uncommon ground: toward reinventing nature. W.W. Norton & Co.: New York.
De Sarra, M. 1935. El abastecimiento de aguas de Barcelona y los yacimientos de sles potsicas. Qumica e
Industria XII (134): 51-56. (Biblioteca de Catalunya)
Duran Ventosa, L. 1914. Memoria sobre alguns problemes ques presenten en el riu Llobregat. Diputaci Provincial
de Barcelona. Barcelona (Biblioteca de Catalunya).
Estevan, A. and N. Prat. 2006. Alternativas para la gestin del agua en Catalua: Una visin desde la perspectiva de
la nueva cultura del agua. Fundacin Nueva Cultura del Agua. Zaragoza and Bakeaz: Bilbao, Spain.
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
Ferret, J. 2006. Els orgens del processos de salinitzaci de les aiges subterrnies de la conca del Llobregat per les
explotacions de sals potssiques (1923-1936). Unpublished manuscript.
Ferret, J. 1985. Laprofitament de les aiges subterrnies del Delta del Llobregat. Comunitat dUsuaris DAiges de
lrea Oriental del Delta del Riu Llobregat. LHospitalet, Spain.
Gorostiza, S., J. Honey-Ross and R. Lloret (in prep). Rius de Sal: Una visi histrica de la salinitzaci dels rius
Llobregat i Cardener durant el segle XX.
Gorostiza, S. 2010. El conflicto salino en el suministro de agua a Barcelona (1925 1940). Encuentro Cientfico
Salud y ciudades en Espaa, 1880-1940. Condiciones ambientales, niveles de vida e intervenciones
sanitarias. 8 y 9 de julio de 2010, Barcelona.
Hughes, R. 1992. Barcelona. Knopf: New York.
Latorre, X. 1995. Historia de laigua a Catalunya. Labecedari. Premia de Mar, Spain.
Lloret, R. 2004. La qualitat de laigua del riu Llobregat. Un factor limitant del passat, un element clau per al futur.
In Prat, N and Tello, E. El Baix Llobregat: histria i actualitat ambiental dun riu. Centre destudis
Comarcals del Baix Llobregat: 92-141.
Lpez, G. 1926. Las Aguas de Barcelona : impugnacin a la memoria de Los Servicios de la Sociedad General de
Aguas de Barcelona. (Biblioteca de Catalunya)
Marsh, G. P. 1874. The Earth as Modified by Human Action. A New Edition of Man and Nature. Sampson Low,
Marston Low, and Searle: London. Republished by Elibron Classics 2006.
Martn Pasqual, J.M. 2007. Aigua i Societat a Barcelona entre les dues Exposicions (1888-1929). Doctoral
Dissertation. Departament dHistria Contempornea i Moderna. Universitat Autnoma de Barcelona.
Available online: http://www.tesisenxarxa.net/TDX-1213107-105345/
Masats i Llover, J. 1997. Histria de la Indstria Txtil a Castellbell i el Vilar. Centre d'Estudis del Bages. Manresa,
Spain.
McCully, P. 2001. Silenced Rivers. The ecology and politics of large dams. Zed Books: New York.

81.
Mooney, H. and P. Ehrlich. 1997. Ecosystem Services: A Fragmentary History. In G. C. Daily, Nature's Services:
Societal dependence on natural ecosystems (pp. 11-19). Island Press: Washington, D.C.
Oliver, B., J.J. Alonso and J.R. Catalan. Estudio Hidrolgico del Ro Llobregat. 1971. CEIA, Litocolor, SA:
Barcelona.
Pedraz Yaez, G. 2007. Proyecto y obra de la Mejora del Tratamiento de Aguas por Osmosis Inversa en la ETAP de
Sant Joan Despi. Documento 1. Memoria. Aguas de Barcelona: Barcelona.
Puig i Valls, R. 1904. El Llobregat: sus cuencas alta, media y baja y obras indispensables que hay que realizar en
ellas, para conseguir que las inundaciones sean cada vez menos temibles, y las aguas normales ms
constantes, con aumentos de riqueza pblica y particular. Memorias de la Real Academia de Ciencias y
Artes de Barcelona, tecera poca IV (40): 524-536. Barcelona.
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pgs 357-366, 377-388, 427-439.
Reguant, J. 1997. Sria: Histria en imatges. 1894-1975. Angle Editorial: Manresa, Spain.
Reisner, M. 1993. Cadillac Desert: The American west and its disappearing water. Penguin Books: New York.
Santalla Torrens, E. 2008. Quan el vapor de la Burs Parl Angls: Impremta Pags: Barcelona.
Schneider, D.W. 1996. Enclosing the floodplain: Resource conflict on the Illinois River, 1880-1920. Environmental
History 1(2): 70-96.
Sorlini, S. and C. Collivignarelli. 2005. Trihalomethane formation during chemical oxidation with chlorine, chlorine
dioxide and ozone of ten Italian natural waters. Desalination 176: 103-111.
Swetnam, T.W., C.D. Allen and J.L. Betancourt. 1999. Applied Historical Ecology. Ecological Applications 9(4):
1189-1206.
Tello, E. and J. Ostos. 2011. Water consumption in Barcelona and its regional environmental imprint: a long-term
history (1717-2008). Regional Environmental Change 1-15.
Voltes i Bou, Pere. 1967. Historia del Abastecimiento de Aguas de Barcelona. Barcelona. Ed. Sociedad General de
Aguas de Barcelona: 198-199.
White, R. 1995. The Organic Machine. Hill and Wang: New York.
Worster, D. 1985. Rivers of Empire. Water, Aridity, and the Growth of the American West. Oxford University
Press: New York.

82.
CHAPTER 3. URBAN ECOSYSTEM SERVICES AND TECHNOLOGICAL
CHANGE
Abstract

Research on ecosystem services has focused mostly on natural areas or remote places, with less attention

given to urban ecosystem services and their relationship with technological change. Perhaps this is

because environmental planners, scientists and managers often view technology a substitute for services

previously obtained from ecosystems. This substitution ostensibly reduces our reliance on natures

services since the superiority of the engineered system motivated the replacement in the first place. I

argue that the expected tradeoff between natural and manufactured capital is false. Rather, the adoption of

new technologies is complementary to ecosystem management. This point is illustrated with a case study

in Barcelona, Spain where the installation of sophisticated water treatment technology increased the value

of ecosystem services found there. This finding suggests that technological change is not a barrier for the

implementation of ideas about ecosystem services. Instead, we can expect the value of ecosystem services

to co-evolve with technological change. I propose that new technologies may shift which ecosystem

functions and structures are valuable but are unlikely to obviate the need for them entirely. New

technologies can generate new opportunities to harness value from ecosystems, and the engineered

structures found in cities may generate more reliance on ecosystem processes, not less.

Keywords: Urban Ecosystem Services, Natural Capital, Technology, Substitution, Water Treatment,
Desalination

83.
3.1 Introduction
Research on ecosystem services has focused mostly on natural areas or remote places that harbor

biological diversity (Naidoo and Ricketts, 2006; Armsworth et al., 2007; Fisher et al., 2011). Less

attention has been given to ecosystem services in human-dominated landscapes (e.g. Goldstein, 2007;

Eigenbrod et al., 2009; Bai et al., 2011). The focus on natural settings is driven by the assumption that

ecosystem services are more likely to emerge from places untouched by human settlement. Urban

environments, by contrast, are rarely associated with ecosystem services, but rather with engineered

structures, artificial materials and technology.

The focus on natural areas makes sense if one remembers that the notion of ecosystem services

was developed by conservation biologists (Ehrlich and Mooney, 1983; Daily, 1997) in collaboration with

economists to find new arguments for ecosystem protection (Costanza et al., 1997). This origin has left an

imprint on the field, as researchers are far more likely to examine ecosystem services in biologically rich

countries, often in the developing world, rather than in industrialized nations (Norgaard, 2010).

Implementing the concept of ecosystem services is primarily being advocated for developing countries

while in the developed countries, with few exceptions, it is much less frequently advocated, let alone

implemented (Norgaard (2010), 1222). Ideas about ecosystem services were well received in the

developing world because they unified conservation and development goals (Sachs and Reid, 2006). It

gave the land owners of biologically rich areas a new reason for stewardship. And the programs that paid

for ecosystem services provided financial incentives for conservation (Wunder, 2008).

This chapter looks at ecosystem services in urban environments and examines the relationship

between technological change and the value of ecosystem services. I question the implicit tradeoff

between the two and find that technological change can increase the value of some ecosystem services. In

the section 2, I motivate the study of ecosystem services in urban environments. Then, in section 3, I

analyze the relationship between ecosystem services and technology and present my argument on how

84.
this relationship may be symbiotic and complementary. Next, in section 4, I provide a case study in which

I examine how new water treatment technologies in Barcelona, Spain altered the mix and value of

ecosystem services found there. The case traces the value of three ecosystem services: water quality

protection, thermal regulation and nutrient cycling before and after technological change. Finally, in

section 5, I extrapolate lessons from the case and discuss the broader implications for research on

ecosystem services in urban environments.

3.2 Urban Ecosystem Services

Relatively few scholars have ventured into urban settings to identify the ecosystem services found

within city limits. At the same time, ecologists have a long history of studying urban ecosystems

(Sanders, 1984; McDonnell and Pickett, 1990; Forman, 1995). Undeterred by the smog, smoke and

cement that dominate most metropolitan areas, urban ecologists have produced a reputable specialization

within ecology that examines how species survive, behave and reproduce in urban environments.

Abandoned lots, train tracks, city parks and parking lots provide unique conditions for ecologists to test

theories about adaptation, selection or island biogeography. And the insight from urban ecology has

generated practical guidelines for managing ecosystems in cities. This study of ecology in cities has

established the foundation for our understanding of ecological processes in urban environments.

Perhaps somewhat surprisingly, urban ecologists have found that metropolitan areas support a

wealth of species diversity, albeit largely non-native, that can surpass the diversity found in rural areas

(Pickett et al., 2011). Biological diversity in urban environments is also strongly correlated with the social

and economic conditions of its human inhabitants (Hope et al., 2003). This co-evolutionary relationship

between ecological and social systems has raised doubts about studying urban ecology in isolation from

the social context. If the ecological system being examined is, in practice, a reflection of the human

conditions that surround it, then perhaps it makes more sense to study the human-environmental system

as a singular entity. This line of research has been pursued by urban ecologists who study the ecology of
85.
cities. In this approach, the human and biophysical environments are analyzed in unison, and the city is

studied as a socio-ecological system that consumes energy, materials, and water. Research on the

ecology of cities most clearly overlaps with the discussion taking place in the literature on ecosystem

services. Notwithstanding, few authors who study the ecology of cities contextualize their research as

one concerning ecosystem services, even though the questions they ask are relevant to disentangle the

complex relationship between ecosystems and human well-being.

Clearly, urban ecosystems provide multiple services for their human inhabitants. Urban ecologists

have documented how urban street trees purify our air, dampen noise and moderate summer heat; while

parks offer places to play, and wetlands process our waste (Forman, 1995; Bolund and Hunhammar,

1999). City street trees in Chicago capture 17 metric tons of C ha-1 y -1 while urban forests in Oakland,

California capture 11 metric tons of C ha-1 y -1 (Pickett et al., 2001). Urban rivers cycle nutrients, and

reduce the threat of flooding (Grimm, 2005; Lundy and Wade, 2011). Many times, the services provided

by particular urban ecosystems are clear. But sometimes, certain areas are overlooked. A review of

ecosystem services in England - a human-dominated landscape - showed that ecosystem services came

from areas that were not targeted for payments for environmental services (Eigenbrod et al., 2009). Thus

the sites that generate the most ecosystem services are not necessarily the ones prioritized for protection.

We also tend to associate urbanization with a decline in ecosystem services. For example, in a

study in San Antonio, Texas, researchers examined the effects of urban sprawl on ecosystem service

provision. They found that urbanization led to a 4% net decline in the value of ecosystem services

because of the conversion of rangelands, forests and cropland to urban areas between 1976 and 1991

(Kreuter et al., 2001). The value of lost ecosystem services were estimated based on $/ha -1 coefficients

per biome type reported in Costanza et al. (1997). Urbanization has also been found to reduce the

capacity of streams to process nutrients (Meyer et al. 2005, Grimm et al. 2005). The suite of services lost

in urban rivers and streams are a central feature of the urban stream syndrome (Walsh et al., 2005).

86.
Urbanization can change the demand for ecosystem services in addition to reducing their supply.

Urban and suburban growth usually increases population densities, increasing the demand for services

related to water purification, flood control, recreation, and food production, among others. A study in the

Leipzig-Halle region of Germany found that between 1990 and 2007, urban growth increased the demand

for ecosystem services, especially in rural and suburban areas (Kroll et al., 2012). Modeling the impact of

future land use changes on the spatial distribution of ecosystem services remains challenging, because

projected land use changes will alter both supply and demand simultaneously. In a study in the United

Kingdom, projected land use changes led to decreases in carbon storage and agricultural services, in

results that were otherwise somewhat inconclusive (Eigenbrod et al., 2011).

Ecosystem services can also be designed into urban environments. Urban planners have taken the

lead in integrating ecological concepts and functions into city landscapes. Frederick Law Olmstead

designed an Emerald Necklace of green areas along the Boston Fens and Riverway. This plan showed

his appreciation for green spaces as useful amenities that can simultaneously provide multiple services for

the public (Spirn, 1984). Later, Ian McHargs seminal piece Design in Nature encouraged planners to

think about how to exploit the physical layout of a site to improve user functionality (McHarg, 1969).

More recently, landscape architects and urban planners are studying how ecological processes can

be integrated into cities with urban designs that include solar panels, green roofs, or rain gardens. Green

roofs improve storm water management, moderate building temperature, reduce the urban heat island

effect, and provide habitat for wildlife (Oberndorfer et al., 2007). Lundy and Wade (2011) propose

designing-in ecosystem services into urban water management through the restoration of urban rivers.

The new language of ecosystem services and green infrastructure is starting to be adopted by

urban planners and city managers even if this new term merely describes practices that were well known

in the past. Indeed, the management of green infrastructure for ecosystem services is emerging as a

major new objective for city managers (Tzoulas et al., 2007). Similarly, a survey of city arborists finds

that managing green areas for their ecosystem services has become a new priority, and that maximizing
87.
these services for local residents is beginning to take precedence over more traditional goals in urban

forestry (Young, 2010).

City planners are enthusiastic about the potential to integrate green infrastructure and ecosystem

services into their designs. However ecosystems are not always benevolent, but can be unpleasant,

dangerous and produce unintended or negative consequences for human welfare. Urban ecosystem

disservices might include allergens, invasive species that eliminate native organisms, the hosting of

pathogens and pests, the obstruction of mobility, and an increase in greenhouse-gases (Pataki et al., 2011).

3.3 Ecosystem Services and Technology


The role of technology is largely absent from discussions on ecosystem services in human-

dominated landscapes. If mentioned at all, engineered systems are perceived as alternatives that compete

with green infrastructure and the services they offer. For advocates of an ecosystem approach,

technology is suspect because innovation may lead decision makers to replace ecosystem functions with

structural solutions.

Technology has always mediated our relationship with ecosystems and the services they provide.

Tools and instruments of various kinds have allowed humans to extract the benefits generated by

ecosystems for centuries. Irrigation techniques helped the ancient Egyptians harness water from the Nile

in order to boost agricultural production sometime around 3000 B.C. (Worster, 1985). Similarly,

mechanical farming has helped increase crop production (Boserup, 1976; Aldy et al., 1998; Raudsepp-

Hearne et al., 2010).

Yet in the literature on ecosystem services, technological change is perceived as a substitute for

natures services. Advocates for ecosystem services have certainly contributed to the juxtaposition

between technology and ecosystems as substitutes. The well known example of water purification

services for New York City (Daily and Ellison, 2002), or wastewater treatment with wetland ecosystems

instead of municipal treatment (Bolund and Hunhammar, 1999) are the clearest illustration of this

88.
substitution. Choosing nature over hard infrastructure has been hailed as financially superior due to lower

capital investments and lower maintenance costs (Chichilnisky and Heal, 1998).

When technology and ecosystems are compared side by side, environmentalists remind us that

ecosystems are more complex and self sustaining (Ehrlich and Mooney, 1983; Costanza, 2003). It has

been shown that technology is an inadequate replacement for ecosystems because engineered systems

lack the multi-functionality and connectivity of natural systems (Moberg and Rnnbck, 2003).

Ecosystems provide many services simultaneously, while a technology is usually designed for only one

(Raudsepp-Hearne et al., 2010). A wetland can treat wastewater and provide species habitat, or offer

recreational opportunities and produce aesthetic value. In an urban context, wetlands, detention ponds

and rain gardens may provide multiple ecological and cultural services (recreational and aesthetic values),

while a manufactured stormwater system is designed for the singular service associated with evacuating

urban runoff. Or when shrimp farmers convert mangroves to pools for aquaculture, they trade several

ecosystem services for the singular service of food production (Barbier et al., 2008). These examples

show us that the direction of the substitution matters greatly. The field of ecosystem services doubts

technologys ability to substitute for ecosystems, and champions ecosystems over technology.

Perhaps the case that has most contributed to our mental juxtaposition of nature versus

technology in the context of ecosystem services was New York Citys decision to rely on watershed

purification services instead of engineered filtration - a decision that saved the city an estimated $4.5

billion, plus $300 million in maintenance expenses (Chichilnisky and Heal, 1998). This celebrated

example thrust the notion of ecosystem services into the spotlight for the research community and popular

press (New York Times, 1994; Guterl, 2005). New Yorks management of ecosystem services generated

the expectation that other municipal governments, or even private businesses, would be able to adopt

similar ecosystem-based strategies with comparable financial outcomes. More than a decade later,

however, these expectations have not been met; nor has the success from New Yorks experience been

replicated in other cities (McCauley, 2006). The reason was that by the late 1990s, most municipal water
89.
suppliers in the United States had already installed the expensive filtration system that New York avoided

(NRC, 2000). With the filtration technology already in place, ecosystem management was no longer

deemed an option. This outcome fed the idea that relying on ecosystem services may be a viable in

pristine contexts, such as the Castskill Mountains in upstate New York, but has a limited potential in

technologically advanced landscapes, such as those municipalities that had already installed the water

filtration systems. This conclusion also captures our intuitive sense that the management of ecosystem

services is appropriate for pristine ecosystems, but once the engineered system is installed; there is no

reason to return to primitive methods of ecosystem management. The superiority of the engineered

system was the reason it was installed in the first place. According to this view, in technologically

sophisticated environments, ecosystem services simply cannot compete.

Thus many assume that our reliance on ecosystem services is temporary until an engineered

alternative is found; one that might be more efficient, cost-effective or both - at which point, resource

users will switch to the designed systems instead. The necessity to switch to engineered systems has

frequently been motivated by the modification or degradation of well-functioning ecosystems. For

instance, engineers developed water treatment technologies in the 19th century because of increasing

pollution in rivers and lakes (Melosi, 1999; Schneider, 2011). In these circumstances, the value of

ecosystem services becomes more salient once they are lost.

However even in circumstances in which ecosystems are functioning adequately, technological

innovation may develop the means to provide superior services. Consider the case in Gramercy,

Louisiana, where a wetland processed wastewater from Zapps Potato Chip Plant. The tertiary waste

water treatment services provided by the wetland were initially valued at $215,220 ($34,700/acre).

However when the volume of waste emitted from the plant exceeded the wetlands treatment capacity, the

company worked with their congressional representative to secure a federal earmark for $150,000 USD to

build a major high-tech waste treatment facility (Sagoff, 2011). In this case, the value of the ecosystem

service was tied to the absence of a particular technology. Once the technological conditions changed, the
90.
wetland could not be valued in the same way, nor be attributed the value of $34,700 per acre. While the

wetland certainly provided other services, those that generated the most value were substituted by the

designed system.

Similarly, technological innovation can motivate a switch away from natural systems to designed

systems. Take for example, the case of almond growers in California, who rely on pollination services

from bees. Nearby, citrus growers keep the same bees away from their crops because cross pollination

causes the citrus fruit to develop undesirable traits. If science were to develop new almonds varieties,

genetically-modified to self-pollinate, this innovation might encourage both almond and citrus farmers to

prefer artificial means of pollination. Thus the value of the pollination ecosystem services for both

growers is situated within its current technological context. Sagoff (2011) expands, The exchange value

of an ecosystem servicelike that of any goodis constantly negotiated in view of market conditions.

Technology also changes. One of the largest nurseries in California, the Dave Wilson Nursery, has

introduced an excellent self-fertile, i.e., self pollinating, almond. If a new cultivar of almond that sets

itself proves profitable, the citrus growers might pay the almond growers to adopt it, to eliminate the

presence of pollination. (Sagoff, 2011)

The possibility that new technology will eventually substitute for ecosystem services has led

authors to criticize the field for being nave, or offering solutions that will be ineffective in the long run.

In a highly cited piece from Nature, McCauley (2006, 28) makes the case that conservation based on

ecosystem services commits the folly of betting against human ingenuity. The entire history of

technology and human progress is one of producing artificial substitutes for what we once obtained

from nature, or domesticating once-natural services I would argue that conservation plans that

underestimate the technological prowess of humans are bound to have short life spans. According to this

view, ecosystems are unlikely to remain competitive with technological improvements because humanity

will always be discovering better ways to meet our needs. These authors emphasize a tradeoff between

91.
services generated by ecosystems and those offered by technological substitutes (McCauley, 2006; Sagoff

2011).

Fossil fuel-based technologies in particular have allowed society to substitute for ecosystem

services, with synthetic fertilizers replacing manure, water treatment systems replacing the capacity for

self purification of rivers and lakes, or engineered structures replacing natural flood and erosion control

processes (Moberg and Rnnbck, 2003; Raudsepp-Hearne et al., 2010). Self proclaimed technological

optimists assert that human progress can be decoupled from ecological conditions entirely, and that this

substitution may continue as long as innovation outpaces environmental degradation (Goeller and

Weinberg, 1976; Small and Jollands, 2006).

In this chapter I argued that the assumed tradeoff between ecosystem services and technology is

false, since these approaches can be complementary. Indeed, technological change can raise the value of

ecosystem services previously ignored, generate a demand for new services, or create new services

entirely. Neoclassical economic theory originally made the simplifying assumption that natural capital

could be substituted for other production inputs including manufactured capital (Gmez-Baggethun et al.,

2010). As evidence of this simplifying assumption, ecological economists often cite the following quote

by Robert Solow (Costanza and Daly, 1992; Clevland and Ruth, 1997; Gmez-Baggethun et al., 2010):

If it is very easy to substitute other factors for natural resources, then there is in principle no problem.

The world can, in effect, get along without natural resources, so exhaustion is just an event, not a

catastrophe (Solow (1974), 11).

However this statement is taken out of context, since Solow himself argues that there are degrees

of substitution, not perfect substitution. A central contribution from ecological economics has been to

note the limits to substitution and the irreplacibility of many forms of natural capital (eg. Costanza and

Daly, 1992; Clevland and Ruth, 1997; Stern, 1997; Gmez-Baggethun et al., 2010). Ecological

economists have pointed out that manufactured capital will not be able to substitute for natural capital

92.
indefinately because our economic system is embedded within the limits of the natural world. It is

unlikely, for example, that manufactured capital will be able to replace lost biodiversity.

Costanza and Daly (1992) were the first to posit that natural and manufactured were

complementary, rather than substitutes: For any given level of technical knowledge, human made capital

and natural capital are, in general, complements, not substitutes (Costanza and Daly, 1992). However

this observation was made against the backdrop of the debate between technological optimists and

skeptics. Therefore the substitution moved in one direction: trading natural capital for manufactured

capital. My argument takes this proposition one step further. Even when sophisticated technology has

already been installed, either to replace ecosystem functions, or isolate managers from the uncertainty

associated with ecosystems, natural capital will remain complementary to technological change. That

is,the complementarity will remain for the very ecosystems that are affected by the technological change.

Furthermore, the mix of services used in the future will be different than today, with some services being

created by the technological change.

Emphasizing that technology and human choice mediates the supply of ecosystem services

amends our common conception of natural capital and the services they provide. Ecosystem services are

not intrinsic functions of the natural environment, fixed, distant and removed from society. Rather the

supply and value of ecosystem services are linked to human choices and technological change, and

therefore in constant evolution. Yet the myth of a fixed supply of ecosystem services, steadily decreasing

or deteriorating, has been enshrined as a central tenet in the field (eg Schrter et al., 2005; MA, 2005; Liu

et at.. 2010). Discussions about a decreasing supply of ecosystem services overlook our fluid relationship

with ecosystem services in the context of technological change. Too frequently we ignore the possibility

that human choice will alter which functions are valuable. Recognizing that the supply of ecosystem

services is not fixed opens the possibility of uncovering new ecosystem services in the future.

Furthermore, the malleable value generated by ecosystems implies that the study of ecosystem services

will remain relevant in technologically advance societies. My argument addresses both technological
93.
optimists and skeptics alike. For the technological optimists, the complementary feature of ecosystems

reaffirms that the benefits of technological progress will ripple out to society in innovative and

unexpected ways. While for the technological skeptics, I highlight the enduring relevance that

ecosystems will have in a technologically advanced future.

In the case study that follows, I examine how a major technological improvement at two drinking

water facilities changed the value of ecosystem services found upstream. I find that technology and

ecosystem services remain complementary even after the membrane system was installed.

3.4 Case Study: Water Treatment Technologies in Barcelona, Spain


The city of Barcelona relies on surface water from the Llobregat River for nearly 50% of its

drinking water supply (Mujeriego, 2006). Two treatment facilities withdraw water from the river. The

Aiges Ter-Llobregat (ATLL) facility in Abrera, Spain is a public water treatment wholesaler that

supplies municipal providers. Downstream, a second water treatment facility is owned and operated by

the private water company Aiges de Barcelona (AGBAR) in Sant Joan Desp. Between 2008 and 2010

both treatment plants simultaneously improved their treatment process by adding new desalination

membranes. ATLL purchased Electrodialysis Reversal (EDR) technology (Valero and Arbs, 2010) while

AGBAR installed reverse osmosis membranes (Luque, 2008). This technological improvement was

motivated by new water quality standards that reduced the permissible level of total trihalomethanes to

below 100 g/L in drinking water (Royal Decree, 2003). In order to comply with the new regulation,

water treatment managers concluded that desalination technology was necessary.

The surface waters from the Llobregat River are mineralized, with an especially high

concentration of salts and bromine (Fernandez-Turiel et al., 2003). These pollutants enter the river

through surface or groundwater flows that come into contact with the large mountains of salt deposited by

an extractive potash industry upstream (Fig 3.1) (Lloret, 2004; ATLL, 2008; Luque, 2008). The cause of

the Llobregats salinization was initially disputed by the mining companies, since the natural geology of

94.
the watershed contains the same minerals polluting the river. However a historical review (Chapter 2) as

well as an analysis using isotope tracers has linked the mineralization of the Llobregat River with

industrial processes from the potash mining industry (Otero and Soler, 2002).

Figure 3.1 Mine tailings in the mid section of the watershed are a major source of salinity in the
Llobregat River. The tailings are large mountains of salt exposed to the open air and in contact with the
groundwater below. Trenches have been built around the perimeter of the salty mountains to collect the
storm water, and channelize the runoff into a brine collector that transports the concentrate to the
Mediterranean. However the mine tailings remain a major source of salinity through groundwater
movements, and several springs surrounding the mine tailings have become saline. The names of the eight
mine tailings are Runam Nou, Runam Vell, Vall Salina (in Cardona), Cabanassas, Fustaret (in Sria),
Vilafruns, El Cogull and La Botjosa (in Sallent).

3.4.1 Advanced Membrane Technologies for Water Treatment

In 2009 the ATLL treatment facility installed an Electrodialysis Reversal (EDR) desalination

system to remove bromine, the critical precursor to trihalomethanes, and ensure compliance with the new

95.
water quality standards. Dissolved ionic bromines are problematic for water treatment managers because

they contribute to the formation of trihalomethanes during the water treatment process (Nokes et al.

1999). Water with high concentrations of bromine and organic matter will form trihalomethanes when

they are mixed with chlorine during treatment (Srinivas Madabhushi 1999, Sorlini and Collivignarelli

2005). The concentrations of trihalomethanes in drinking water supplies are regulated because they are

carcinogenic (Nokes et al. 1999, Villanueva-Belmonte 2003), and higher water temperatures accelerate

the formation of trihalomethanes (Toroz and Uyak 2005). Disinfection by-products such as

trihalomethanes are usually removed by modifying the chlorine process or removing the chlorine

sensitive compounds (Gopal et al. 2007). Together with an expansion in treatment capacity, the new

system cost over 61 million (Valero and Arbs 2010). EDR technology separates dissolved ions such as

Br-, Cl-, and Na+ by applying electrical charges. The membranes reverse polarity between positive and

negative charges every twenty minutes to clean the membranes and reduce fouling incidents (AWWA

1995). When the EDR system was installed it was the largest of its type in the world (personal

communication, Valero 2009).

Simultaneously, the second water treatment plant managed by AGBAR installed an ultrafiltration

and reverse osmosis system at their facility in Sant Joan Desp. Reverse osmosis has been hailed as the

ultimate barrier within the field of water treatment (Greenlee et al. 2009). Its advanced non-porous

material is capable of removing all dissolved solids including pharmaceutical compounds and pesticides

(Elimelech and Phillip 2011). AGBARs reverse osmosis system cost 75 million to purchase and install

(Luque 2008). Reverse osmosis technology is rarely used to treat surface water because these sources

provide lower quality feed water, which cause membranes to foul, and because of the expenses associated

with brine disposal at inland sites (Greenlee et al. 2009). A literature review has failed to uncover any

similar reverse osmosis systems treating surface water for municipal use. Only industrial users needing

premium water quality have installed reverse osmosis for treating surface water along the Main River in

Germany (Clever et al. 2000). On the other hand, reverse osmosis is frequently used to treat brackish
96.
groundwater (Greenlee et al. 2009). In comparison with older technologies, reverse osmosis marks a

clear departure from the traditional municipal water treatment systems that dominated the 20th century.

While EDR and reverse osmosis technologies differ, both water treatment plants share several

features. First, and most obviously, both systems adopt advanced desalination membrane systems that are

expensive to install and operate. Second, the new membranes did not replace the older system, but were

appended onto the pre-existing chain of treatment technologies (Fig 3.2 and Fig 3.3). Traditional water

treatment methods clarify turbid waters with coagulants, flocculants, settlement tanks, and sand and

activated carbon filters, and then disinfect the water with chlorine, chloramine, ultraviolet rays or ozone

(Crittenden et al. 2005). These traditional methods now serve as pre-treatment steps. Both systems were

also designed so that only a percentage of the treated water would be desalinated. Water managers control

the mixing ratio by adding desalination modules. AGBAR seeks a constant mixing ratio of 50%, while

ATLL varies their product mix based on seasonal considerations. Last, and most importantly for our study

of ecosystem services, the new treatment technologies significantly altered the cost structure of the

treatment process.

The new membrane systems incorporated two new water quality parameters into the treatment

cost function: (1) salinity, as measured by conductivity (S/cm), and (2) temperature (C). Both of these

water quality parameters became critical values once the new water treatment systems began to operate.

97.
Figure 3.2 Treatment Process at the ATLL-Abrera facility.

Figure 3.3 Treatment Process at the AGBAR-Sant Joan Desp facility.

By choosing to install new water treatment systems downstream, water managers revealed their

preference for technological strategies over ecosystem approaches for improving source water quality.

And with the new sophisticated treatment system, they attained the capacity to meet output water quality

standards across a much wider range of input water qualities. Therefore, according to conventional

wisdom, the sophisticated new technology further isolated managers from the influence of ecosystem

processes. To assess the impact of technological change on the value of ecosystem services I analyzed

three ecosystem services: water quality protection (affecting salinity); thermal regulation (affecting

temperature); and nutrient cycling (affecting ammonium), before and after the adoption of new membrane

treatment.

98.
3.4.2 The Value of Ecosystem Services Following Technological Change
3.4.2.1 Salinity
The new water treatment technology was adopted to reduce the uncertainty surrounding

freshwater provision, and yet this installation created new linkages with ecosystem functions. Prior to the

installation of the desalination systems, treatment costs were impervious to fluctuations in salinity, largely

because the traditional treatment methods were incapable of removing chloride ions from the feed water.

Water managers did not modify their treatment process in response to fluctuating salinity values.123

However following the installation of the new membranes, salinity became a significant driver of

treatment costs.

Reverse Osmosis Technology at AGBAR-Sant Joan Desp

For reverse osmosis systems, the cost of removing salts is governed by the laws of osmotic

pressure. I estimated reduced treatment costs associated with potential reductions in salinity in the

Llobregat River with the Reverse Osmosis System Analysis (ROSA) software (version 7.2.7), created by

Dow Chemical, the membrane manufacturer. ROSA allowed me to configure a desalination system with

the specification of AGBARs treatment plant in Sant Joan Desp (Appendix A). I ran ROSA with

different concentrations of TDS and observed how much energy was needed to maintain the same system

performance. As expected, higher concentrations of salt required more energy per cubic meter of water

produced (Fig 3.4). The output relationship is nearly linear, but is best fit by an exponential functional

form. When the feed water has higher salinity concentrations, marginal water quality improvements

produce more energy savings than the same marginal improvements in cleaner water. The Llobregat

123
The exception was when salinity values passed maximum permissible concentrations, in which case, surface
water treatment was stopped altogether. These high concentrations usually occurred when the brine collector that
transports mining effluents from the mines to the Mediterranean would rupture and release highly concentrated salt
water directly into the river.
99.
River has an average conductivity of 1500 S/cm. At this reference point, a reduction of 100 S/cm

would save AGBAR approximately 159,612 per year.

Conductivity - Energy Relationship for the


LE-440i Reverse Osmosis Membranes
0.90
0.80 y = 0.315e0.00034x
Specific Energy (kWh/m3)

R = 0.9969
0.70
0.60
0.50
0.40
0.30
0.20
0.10
0.00
0 200 400 600 800 1,000 1,200 1,400 1,600 1,800 2,000
Conductivity (S/cm)

Figure 3.4. The relationship conductivity (S/cm) and specific energy efficiency (kWh/m3) for the LE-
440i reverse osmosis membranes produced by Dow Chemical. Y=0.315e0.00034x R2=0.9969

Electrodialysis Reversal Technology at ATLL-Abrera

The installation of new membrane technologies produced similar changes to treatment costs at the

ATLL treatment plant. Lower salinity levels in the Llobregat River would allow ATLL to produce the

same quality drinking water with less electrical current running through the EDR modules. To quantify

the potential savings associated with salinity reductions I plotted observed conductivity values and energy

use (kWh/m3) at nine EDR modules operating in November 2010 and July 2010 (n=247). For every unit

increase in conductivity (S/cm) the treatment system consumed an additional 0.0002 kWh/m3 (Fig 3.5).

This implies that a reduction in conductivity of 100 S/cm is associated with an increase in energy use of

0.02 kWh/m3.

ATLL annually produces approximately 30 million cubic meters of drinking water with the EDR

system. The energy expenses associated with this production is approximately 939,600 per year with a

100.
mean energy expenditure of 0.348 kWh/m3 and an average energy cost of 0.09 /kWh. Under these

conditions, a reduction in conductivity by 100 S/cm would generate savings of 54,000 per year for the

ATLL treatment plant (Appendix B).

Figure 3.5 The observed relationship between energy consumed (kWh/m3) and conductivity (S/cm) in
the feed water for the EDR water treatment system operated by ATLL. Y=0.0002x + 0.1481. R2 =
0.2586

Water quality protection

The installation of the membrane technology has realigned economic incentives for both water

treatment facilities. With the new desalination systems, water treatment managers now have an incentive

to invest in ecosystem structures or functions that can protect the river from salt pollution. One approach

for reducing salinity concentrations is to restore the mountains of mine tailings with halophytic

vegetation. A restored ecosystem covering the salt mountains would help protect the water quality of the

Llobregat River by reducing the volume of water coming into contact with the salt deposits and filtering

into the groundwater. Restoring vegetation on these salt mountains would require specialized techniques,

since most plant life is anathema to salty soils. However plant biologists have studied halophyte species

which are tolerant to saltier soils and may be candidates for restoration (Adams et al. 1998, Bohnert and

Cushman 2000). Despite inhospitable soil conditions, we can see evidence that the salt deposits can be

restored with living vegetation. In the 1960s, environmental groups and the mining company collaborated

101.
to cover the abandoned Botjosa salt tailings with soil and seeds. The initial planting had difficulties, but

over time vegetation has survived and the results of this project remain visible today (Fig 3.6). The

Catalan Water Agency has also begun to cover one mine tailing at Vilafruns. The agency is covering the

salt mountains with impervious plastics, soils and vegetation in order to reduce the flow of water filtering

through the salt mountains and entering the Llobregat River through groundwater flows (ACA, 2009).

The water agency will be monitoring the project to assess if the restoration is effective at reducing salinity

concentrations in the Llobregat River.

Figure 3.6 A segment of the Botjosa mine tailing in Sallent with restored vegetation growing on top.
(Photo by Author. July 2008)

Based on the salinity-energy relationships presented above, we can estimate the benefits that such

restoration projects could produce for the water treatment managers downstream. For example, given

current conductivity of 1500 S/cm, a reduction in mean daily conductivity values in the Llobregat River

by 100 S/cm would generate ecosystem services worth 213,612 per year. A reduction of conductivity

by 500 S/cm would generate services worth 868,546 per year (Table 3.1).
102.
Table 3.1 Treatment costs savings associated with reductions in conductivity (S/cm) at both treatment
plants with their respective desalination technologies. Conductivity reductions depart from a baseline
value of 1,500 S/cm.
Mean daily
reductions in feed
water conductivity Reduced Treatment Reduced Treatment
(S/cm) Cost AGBAR Cost ATLL Total Savings per year
100 159,612 54,000 213,612
200 272,321 108,000 387,321
300 399,030 162,000 561,030
400 518,740 216,000 734,740
500 598,546 270,000 868,546

3.4.2.2 Temperature
Stream temperature is another water quality parameter inserted into the cost function following

the change in water treatment technology. Because trihalomethanes form more rapidly during warmer

months, stream temperature is particularly important for the ATLL facility. Water managers have

adopted a treatment protocol in which additional EDR modules are turned on as stream temperatures rise

in the spring and summer (ATLL, 2008). Every additional EDR module costs approximately 1,000 per

day to operate (personal comment Valero, 2010).

Stream temperature also controls critical ecosystem processes such as the metabolic rates of

aquatic species (Acua and Tockner, 2009). Warm temperatures also reduce the solubility of oxygen, and

low concentrations of dissolved oxygen make waters less hospitable to fish and invertebrates (Graczyk

and Sonzogni, 1991). Thermal heating is symptomatic of urbanized watersheds and disturbed river

systems (Webb et al., 2008). Stream temperatures can be moderated with the restoration of riparian forest

or increases in stream discharge (Bartholow, 1991).

103.
Thermal Regulation

ATLLs water treatment protocols for the EDR system relies on stream temperature to guide

operation decisions. This choice created a link between the ecological condition of the Llobregat River

and operating costs at the treatment plant. In this case, the sensitivity of the EDR system to high

temperatures increased the value of thermal regulation services offered by riparian forests since the

temperature reductions provided by riparian habitat would allow managers to use less intensive treatment.

The restoration of additional riparian vegetation could further reduce the days in which EDR modules

operate (Chapter 4). Depending on the extent of the forest restoration, the restoration of riparian

vegetation could provide ecosystem services in the range of 57,000-156,000 per year (Chapter 4). This

specific value generated by the riparian forests would not exist in the absence of the EDR treatment.

Increasing stream discharge and restoring floodplains could have a similar cooling effect on stream

temperatures, and reduce treatment expenses for the ATLL facility. In a scenario that would increase

discharge by only 25%, the treatment plant benefitted from ecosystem services valued at 40,000 per

year. Similar to the circumstances found with salinity, the adoption of the EDR technology created a new

incentive for ATLL managers to explore ecosystem management practices that would restore ecological

functions and reduce thermal heating in the Llobregat River.

3.4.2.3 Ammonium

The marginal value of nutrient cycling services may also increase following the installation of

new treatment technologies. The AGBAR-Sant Joan Desp water treatment facility is frequently forced to

halt treatment when the Llobregat River fails to meet minimum water quality standards (Lloret, 2004).

Ammonium concentrations are especially problematic because they often exceed the maximum

permissible values for treatable water. Between 2000 and 2010, high ammonium concentrations forced

104.
AGBAR to stop on 285 occasions (Fig 3.7). Stoppage events generate a penalty cost because they oblige

the treatment company to purchase water at a higher cost elsewhere.

Figure 3.7 Ammonium concentrations at the AGBAR-SJD water treatment facility from 1-January-2000
to 31-December-2010. The line at 2.5 mg/L marks the treatment stoppage threshold. Total exceedance
days in 11 years: 286 (7.1%). (Data AGBAR-SJD)

Ammonium is released into the Llobregat River and its tributaries primarily by wastewater

treatment plants. The concentrations rise during wet weather events because of combined sewer

overflows that release untreated wastewater into the river. Ammonium concentrations are also high during

winter months when the colder temperatures slow down the bacterial nitrification of ammonia

(Mugeriego, 2006).

With the new membrane treatment technology, AGBAR has the ability to treat water with higher

ammonium concentrations. Thus, treatment managers have been studying the possibility of increasing the

stoppage threshold for ammonium. Currently the stoppage threshold is at 2.5 mg/L, but with the new

membrane system, it is feasible to continue water treatment at higher concentrations and thereby avoid

stoppage events. It thus appears that the new technology could substitute for ecosystem services that

reduce ammonium concentrations in the watershed.


105.
Nutrient Cycling

Ecosystem processes play a critical role in determining the concentration of ammonium in the

Llobregat River. Nitrification occurs in the rivers soils or other surfaces where biofilms accumulate

(Butturini et al., 2000), and streams that are impaired have lower nitrification rates than rivers in good

ecological condition (Mart et al., 2004). Restoration measures such as recovering meanders, widening

flood plains, hyporheic restoration, or revegetation can increase instream nitrification rates (Admiraal and

Botermans, 1989; Butturini et al., 2000; Kaushal et al., 2008). In the case of the Llobregat, river

restoration measures would be particularly valuable because higher nitrification rates could also reduce

the number of days in which water treatment is stopped and penalty costs are assumed.

Under these new circumstances, the value of investing in nitrification services may increase or

decrease, depending on where the new threshold is located (Fig 3.8). Unexpectedly, once a new

technology has allowed water treatment managers to raise the stoppage threshold to a higher level, the

same investment in ecosystem services might be worth more with the technology than without. Take for

example, an ecosystem service that reduces ammonium concentrations by an average of 4% (Fig 3.8).

This same service has a higher marginal value if the new technology allowed the stoppage threshold to be

revised at around 2.8 mg/L than it would at lower threshold levels. In this case, improved nutrient cycling

would produce greater marginal benefits than that same investment would have produced under previous

conditions, before the new technology (Brozovic et al. in prep). This makes intuitive sense if one can

imagine the new (higher) stoppage treatment threshold being located just below a cluster of peak

ammonium values.

106.
Figure 3.8 The value of ecosystem services related to the nitrification of ammonium (z-axis) given
different stoppage threshold levels (x-axis) and different percent reductions in ammonium concentrations
(y-axis). Figure from (Brozovic et al. in prep)

3.5 Discussion

Urban environments provide new opportunities to discover and manage ecosystem services. The

engineered structures found in cities may generate more reliance on ecosystem processes, not less. In the

case of water treatment in Barcelona, the new membrane technology did not eliminate the demand for

ecosystem services associated with water purification. To the contrary, several ecosystem services that

regulated water quality became more valuable. Water treatment managers installed the new membrane

systems to allow for more continuous treatment across a wider range of input water qualities in essence,

protecting themselves from ecosystem variability and uncertainty. And yet unexpectedly, the new

technology created additional reasons to invest in ecosystem management.

107.
Considerable attention has been given to how technology can liberate society from the constraints

of the natural world. In water treatment, filtration systems reduced the value of watershed services

associated with soil and sediment retention, and chlorination reduced the value of natural processes that

eliminated infectious bacteria. These technological advancements have created unquestionable benefits

for society. However less attention has been directed at how ecosystem management may complement

the new technologies once they already have been adopted.

Of course, technological change will not always increase the value of ecosystem services. Recall

the case from Gramercy, Louisiana, where the wetlands lost much of their value following the installation

of a wastewater treatment facility, or in California, where farmers might switch to genetically modified

varieties that eliminate the need for pollination services. As our technology changes, so too will the values

we assign to different ecosystem processes. In a more technologically advanced future, the bundle of

ecosystem services we use will transform as different technologies favor different sets of ecosystem

functions. Ecosystem values are not static. Rather they are a function of human preferences and our

interaction with technology (Pritchard et al., 2000).

This case also invites a more controversial claim: that technological change can create new

ecosystem services. In Barcelona I observed how new services were generated from existing ecosystem

structures and functions because of technological choice. Certainly, the ecosystem structures and

functions that protected water quality and reduced stream temperatures existed prior to the membrane

technology, but without the membrane technology, society did not benefit in the same way, nor were the

values of ecosystem services as connected to explicitly defined beneficiaries. While these ecosystem

structures and functions produced other valuable services (nutrient cycling, flood mitigation, recreation,

etc...) the specific services produced did not exist prior to the membrane treatment. In this case I focus on

use values, but there is no reason why this argument cannot be generalized to ecosystem services that

generate non-use values. For example, underwater exploration technologies may allow us to view and

108.
appreciate submerged life forms previously unknown or unseen, increasing their value to society. Even in

the context of non-use values, technology may increase the value of ecosystems and their services.

Barcelonas experience in water treatment suggests that ideas about ecosystem services remain

relevant in technologically advanced settings. Globally, over 1 billion people remain without clean

drinking water and approximately 2.3 billion live in water scarce regions (Service, 2006). If countries in

the developing world leap-frog to advanced water treatment systems to purify their contaminated sources,

the management of ecosystem services must be considered simultaneously as a viable option to improve

river ecology and reduce water treatment costs. Even in technologically sophisticated environments, an

ecosystem services approach can uncover new management options that may bring both environmental

and economic benefits.

This chapter has sought to underscore the potential of ecosystem management in urban and

technologically advanced contexts. The new system allowed water managers to improve drinking water

quality and increased water security. It is difficult to imagine ecosystems providing equivalent

improvements in quantity, quality and certainty in such a short period of time. However the water quality

improvements from the new technological systems also came with a considerable financial and

environmental cost. Addressing water demands with energy-intensive systems is unlikely to be a

sustainable solution in the long run (Mehan, 2009). Instead, water managers would do well to explore

ecosystem alternatives regardless of their technological circumstances. The mere presence of modern

treatment systems should not blind us to opportunities to manage ecosystem services.

This chapter has shown that the installation of new technologies is not a barrier for the application

of ideas about ecosystem services. While technology change may obviate the need for some ecosystem

services, new technology will also generate a demand for new services or previously unvalued services.

The values ascribed to ecosystems will depend on the context and the technology that surrounds it.

Future technologies will rearrange the importance that managers will place on different environmental

parameters and services. The rise of new technologies will not reduce our reliance on ecosystem services,
109.
but rather ecosystems will benefit us in new and different ways. Regardless of the technological

environment the future may hold, investing in knowledge about ecosystem services will remain a wise

development strategy.

3.6 Acknowledgements

This research has received funding from the Centre Tecnolgic de lAigua (CETaqua) and the Agncia

Catalana de lAigua (ACA). I thank AGBAR and ATLL for providing water treatment data and Benoit

Lefvre for assistance with the ROSA model.

110.
3.7 References Chapter 3
Acua, V. Tockner K., 2009. Surface-subsurface water exchange rates along alluvial river reaches control the
thermal patterns in an Alpine river network. Freshwater Biology 54, 306-320. Doi:10/1111/j.1365-
2427.2008.02109.x
Adams, P., Nelson, D.E., Yamada, S., Chmara, W., Jensen, R.G., Bohnert, H.J., Griffiths. H., 1998. Tansley Review
No 97 Growth and development of Mesembryanthemum crystallinum (Aizoaceae) New Phytologist 138,
171-190.
Admiraal, W., Botermans Y.J.H., 1989. Comparison of nitrification rates in three branches of the lower Rhine.
Biogeochemistry 8, 135-151.
Agncia Catalana de lAigua (ACA) 2009. Projecte constructiu de les actuacions desdinades a la reducci de
limpacte ambiental del runam inactiu de Vilafruns. Departament de Medi Ambient i Habitatge, Generalitat
de Catalunya.
Aldy, J.E., Hrubovcak, J., Vasavada U., 1998. The role of technology in sustaining agriculture and the environment.
Ecological Economics 26, 81-96.
Armsworth, P.R., Chan, K.M.A., Daily, G.C., Ehrlich, P.R., Kremen, C., Ricketts, T.H., Sanjayan, M.A., 2007.
Ecosystem-service science and the way forward for conservation. Conservation Biology 21(12): 1383-4.
Arrow, K., Daily, G., Dasgupta, P., Levin, S., Maler, K.G., Maskin, E., Starrett, D., Sterner, T., Tietenberg, T., 2000.
Managing Ecosystem Resources. Environmental Science and Technology. 34, 1401-1406.
American Water Works Association (AWWA) 1995. Electrodialysis and Electrodialysis Reversal. AWWA Manual
M38. Denver, CO.
ATLL 2008. Estudi per a loptimitzaci econmica-sanitria del funcionament conjunct del tractament convencional
I la instal.laci delectrodilisis reversible a lETAP del Llobregat (T.M. Abrera). Planificaci del servei de
producci daigua per al consum hum. Aiges Ter Llobregat. 17 September 2008.
Bai, Y., Zhuang, C., Ouyan, Z., Zheng, H., Jiang B., 2011. Spatial characteristics between biodiversity and
ecosystem services in a human-dominated watershed. Ecological Complexity 8, 177-183.
Barbier, E.B., 2007. Valuing ecosystem services. Economic Policy 22 (49), 177-229.
Barbier, E.B., Koch, E.W., Silliman, B. R., Hacker, S. D., Wolanski, E., Primavera, J., Granek, E. F. et al. 2008.
Coastal ecosystem-based management with nonlinear ecological functions and values. Science 319, 321-3.
Bartholow, J.M., 1991. A Modeling Assessment of the Thermal Regime for an Urban Sport Fishery. Environmental
Management 15 (6), 833-845.
Bolund, P., Hunhammar, S. 1999. Ecosystem services in urban areas. Ecological Economics 29 (2), 293-301.
Boserup, E., 1976. Environment, population, and technology in primitive societies. Population and Development
Review 2, 21-36.
Bohnert, H.J., Cushman, J.C., 2000. The ice plant cometh: Lessons in abiotic stress tolerance. Journal of Plant
Regulation 19, 334-346
Brozovic, N., Honey-Ross, J., Schneider, D.W. (in prep). Technological Change and the Value of Ecosystem
Services.
Butturini, A., Battin, T.J., Sabater F., 2000. Nitrogen in stream sediment biofilms: the role of ammonium
concentrations and DOC quality. Water Resources 34 (2), 629-639.
Chan, K.M.A., Shaw, M.R., Cameron, D.R., Underwood, E.C., Daily, G.C., 2006. Conservation planning for
ecosystem services. PLoS Biology 4 (11), 2138-2152.
Chan, K.M.A., Hoshizaki, L., Klinkenberg, B., 2011. Ecosystem services in conservation planning: Targeted
benefits vs. Co-Benefits or Costs? PLoS ONE 6 (9):e24378. doi:10.1371/journal.pone.0024378.
Chen, N., Li, H., Wang, L., 2009. A GIS-based approach for mapping direct use value of ecosystem services at a
country scale: Management implications. Ecological Economics 68 (11), 2768-2776.
Chichilnisky, G., Heal, G., 1998. Economic returns from the biosphere. Nature 391, 629-630.
Clever, M., Jordt, F., Knauf, R., Rabiger, N., Rudebusch, M., Hilder-Scheibel, R., 2000. Process water production
from river water by ultrafiltration and reverse osmosis. Desalination 131, 325-336.
Clevland, C.J., Ruth, M., 1997. When, where and by how much do biophysical limits constrain the economic
process? A survey of Nicholas Georgescu-Roegens contribution to ecological economics. Ecological
Economics 22 (3), 203-223.
Costanza, R., Daly, H.E., 1992. Natural Capital and Sustainable Developpment. Conservation Biology 6 (1), 37-46.
Costanza, R., et al. 1997. The value of the worlds ecosystem services and natural capital. Nature 387, 253-260.
111.
Costanza, R., 2003. A vision of the future of science: reintegrating the study of humans and the rest of nature.
Futures 35, 651-671.
Crittenden, J.C., Trussell, R.R., Hand, D.W., Howe, K.J., Tchobanoglous, G., 2005. Water treatment: principles and
design. John Wiley & Sons, Hoboken, New Jersey.
Daily, G. C., 1997. Natures Services. Island Press. Washington DC.
Daily, G. C., Ellison, K., 2002. The New Economy of Nature. Island Press. Washington D.C.
Dow Chemical. 2011. Reverse Osmosis System Analysis (ROSA) 7.2.7. Available online
[http://www.dowwaterandprocess.com/support_training/design_tools/rosa.htm]
Ehrlich, P.R., Mooney, H. A., 1983. Extinction, Substitution, and Ecosystem Services. Bioscience 33, 248-254.
Elimelech, M., Phillip W.A., 2011. The future of seawater desalination: energy, technology and the environment.
Science 333 (6043), 712-717.
Eigenbrod, F., Bell, V.A., Davies, H.N. , Heinemeyer, A., Armsworth, P.R., and Gaston, K. J., 2011. The impact of
projected increases in urbanization on ecosystem services. Proceedings of the Royal Society B 278 (1722),
3201-3208.
Eigenbrod, F., Anderson, B.J., Armsworth, P.R., Heinemeyer, A., Jackson, S. F., Parnell, M., Thomas, C. D.,
Gaston, K. J., 2009. Ecosystem service benefits of contrasting conservation strategies in a human-
dominated region. Proceedings of the Royal Society B. 276 (1669), 2903-2911.
Farber, S., Costanza, R., Childers, D.L., Erickson, J., Gross, K., Grove, M., Hopkinson, C. S., Kahn, J., Pincetl, S.,
Troy, A., Warren, P., Wilson, M. 2006. Linking Ecology and Economics for Ecosystem Management.
Bioscience 56 (2), 121-133.
Fernandez-Turiel, J.L., Gimeno, D., Rodriguez, J.J., Carnicero, M., Valero F., 2003. Spatial and Seasonal Variations
of water quality in a Mediterranean catchment: The Llobregat River (NE Spain). Environmental
Geochemistry and Health 25 (4), 453-474.
Fisher, B., Turner, R. K., Morling, P., 2009. Defining and classifying ecosystem services for decision making.
Ecological Economics 68 (3), 643-653.
Fisher, B. et al. 2011. Measuring, modeling and mapping ecosystem services in the Eastern Arc Mountains of
Tanzania. Progress in Physical Geography 35 (5), 595-611.
Graczyk, D.J., Sonzogni, W.C., 1991. Reductions of dissolved-oxygen concentrations in Wisconsin streams during
summer runoff. Journal of Environmental Quality 20 (2), 445-451.
Goeller, H.E., Weinberg, A.M., 1976. The Age of Substitutability. Science 20, 683-689.
Goldstein, J.H., 2007. Paying for Conservation in Human Dominated Landscapes. Doctoral Dissertation. Stanford
University.
Gmez-Baggethun, E., de Groot, R., Lomas, P.L., Montes, C., 2010. The history of ecosystem services in economic
theory and practice: From early notions to markets and payment schemes. Ecological Economics 69:1209-
1218.
Gopal, K., Swarupa Tripathy, S., Bersilon, J.L., Prabha Dubey, S., 2007. Chlorination by products their
toxicodynamics and removal from drinking water. Journal of Hazardous Materials 140, 1-6.
Guterl, F., 2005. Investing in Green. Newsweek June 1. 36.
Greenlee, L.F., Lawler, D.F., Freeman, B.D., Marrot, B., Moulin, P., 2009. Reverse Osmosis desalination: Water
sources, technology, and todays challenges. Water Research 43, 2317-2348.
Hope, D., Gries, C., Zhu, W., Fagan, W.F., Redman, C.L., Grimm, N.B., Nelson, A.L., Martin, C., Kinzig, A.,
2003. Socioeconomics drive urban plant diversity. Proceedings of the National Academy of Sciences 100
(15), 8788-8792.
Howarth, R.B., Farber, S., 2002. Accounting for the value of ecosystem services. Ecological Economics 41, 421-
429.
Kareiva, P., Watts, S., McDonald, R., Boucher, T., 2007. Domesticated Nature: Shaping Landscapes and
Ecosystems for Human Welfare. Science 316, 1866-69.
Kaushal, S.S., Groffman, P.M., Mayer, P.M., Striz, E., Gold, A.J. 2008. Effects of stream restoration on
denitrification in an urbanized watershed. Ecological Applications 18 (3), 789-804.
Kroll, F., Mller, F., Haase, D., Fohrer, N., 2012. Rural-urban gradient analysis of ecosystem services supply and
demand dynamics. Land Use Policy 29 (3), 521-535.
Liu, S., Costanza, R., Farber, S., Troy, A., 2010. Valuing ecosystem services: Theory, practice, and the need for a
transdisciplinary synthesis. Annals of the New York Academy of Sciences 1185, 54-78.

112.
Lloret, R., 2004. La qualitat de laigua del riu Llobregat. Un factor limitant del passat, un element clau per al futur.
In Prat, N and Tello, E. El Baix Llobregat: histria i actualitat ambiental dun riu. Centre destudis
Comarcals del Baix Llobregat. p. 92-141.
Luque, F., 2008. Tratamiento del Agua del Ro Llobregat en la ETAP de Sant Joan Desp (Barcelona) por
Membranas de Ultrafiltracin y smosis Inversa. Associacin Espaola de Desalacin y Reutilizacin
(AEDyR) VII Congreso AEDyR. December 3-5.
Mart, M., Aumatell, J., God, L., Poch, M., Sabater, F., 2004. Nutrient Retention Efficiency in Streams Receiving
Inputs from Wastewater Treatment Plants. Journal of Environmental Quality 33, 285293.
McCauley, D.J., 2006. Selling out on nature. Nature 443 (27-28), 27-28.
McHarg, I., 1969. Design with Nature. Garden City, NJ. Doubleday/Nat. Hist.
Mehan, G.T., 2009. Congressional testimony before the Subcommittee on Water Resources and Environment of the
House Committee on Transportation and Infrastructure on Sustainable Water Management. 4 February.
Melosi, M.V., 1999. The Sanitary City: City: Urban Infrastructure in America from Colonial Times to the Present.
The Johns Hopkins University Press, Baltimore, MD.
Millennium Ecosystem Assessment (MA). 2005. Ecosystems and human well-being: Current states and trends.
Washington DC: Island Press.
Moberg, F., Rnnbck, P., 2003. Ecosystem services of the tropical seascape: interactions, substitutions and
restoration. Ocean & Coastal Management 46 (1-2), 27-46.
Mujeriego, R., 2006. Abastament daigua des del Baix Llobregat nord: Diagnosi per a la millora de la qualitat.
Agncia Catalana de lAigua, Aigues Ter-Llobregat, Direcci General de Salut Pblica. Generalitat de
Catalunya.
Naidoo, R., Ricketts, T.H., 2006. Mapping the economic costs and benefits of conservation. PloS Biology 4 (11),
2153-2164. e360 DOI:10.137/journal.pbio.0040360
National Research Council, 2000. Watershed management for potable water supply: Assessing the New York City
strategy. Washington, D.C.: National Academy Press.
New York Times, 2004. Save the watershed. Editorial Desk. Sunday Edition Section 4; Page 14; Column 1.
Nikolaou, A.D., Golfinopoulous, S.K., Arhonditsis, G.B., Kolovoyiannis, V., Lekkas, T.D., 2004. Modeling the
formation of chlorination by-products in river waters with different quality. Chemosphere 55, 409-420.
Norgaard, R.B. 2010. Ecosystem services: From eye-opening metaphor to complexity blinder. Ecological
Economics 69 (6), 1219-1227.
Otero, N., Soler A. 2002. Sulphur isotopes as tracers of the influence of potash mining in groundwater salinisation in
the Llobregat Basin (NE Spain). Water Research 36 (16), 3989-4000.
Pritchard, L., Folke, C., Gunderson L., 2000. Valuation of Ecosystem Services in Institutional Contexts. Ecosystems
3 (1), 36-40.
Raudsepp-Hearne, C., Peterson, G.D., Teng, M., Bennett, E.M., Holland, T., Benassaiah, K., MacDonald, G.K.,
Pfeifer, L., 2010. Untangling the Environmentalists Paradox: Why is human well-being increasing as
ecosystem services degrade? BioScience 60 (8), 576-589.
Royal Decree, 2003. Establishment of Spanish Drinking Water Quality Standards. 2003/140. 7 February 2003.
Sanders, R.A., 1984. Some determinants of urban forest structure. Urban Ecology 8 (1-2), 13-27.
Sachs, J.D., Reid, W.V., 2006. Investments toward sustainable development. Science 312 (5776), 1002.
Sagoff, M., 2011. The quantification and valuation of ecosystem services. Ecological Economics 70 (3), 497-502.
Schneider, D.W. 2011. Hybrid Nature: Sewage Treatment and the Contradictions of the Industrial Ecosystem. MIT
Press. Cambridge, MA.
Schrter, D. et al. 2005. Ecosystem Service Supply and Vulnerability to Global Change in Europe. Science 310
(5752), 1333-1337.
Service, R.F., 2006. Desalination freshens up. Science 313 (5790), 1088-1090.
Simpson, D.R., 2001. A Note of the Valuation of Ecosystem Services in Production. Resources for the Future
Discussion Paper. 01-16. Online: http://www.rff.org/documents/RFF-DP-01-16.pdf
Small, B., Jollands, N., 2006. Technology and ecological economics: Promethean technology, Pandorian potential.
Ecological Economics 56 (3), 343-358.
Spirn, A.W., 1984. The Granite Garden: Urban Nature and Human Design. Basic Books: New York.
Srinivas Madabhushi, B., 1999. What are trihalomethanes? On Tap. (Spring), 18-19.

113.
Sorlini, S., Collivignarelli, C., 2005. Trihalomethane formation during chemical oxidation with chlorine, chlorine
dioxide and ozone of ten Italian natural waters. Desalination 176 (1-3), 103-111.
Stern, D.I., 1997. Limits to substitution and irreversibility in production and consumption: A neoclassical
interpretation of ecological economics. Ecological Economics 21 (3), 197-215.
Toroz, I., Uyak V., 2005. Seasonal variations of trihalomethanes (THMs) in water distribution networks of Istanbul
City. Desalination 176 (1-3), 127-141.
Turner, R. K., Paavola, J., Cooper, P., Farber, S., Jessany, V. Georgion, S. 2003. Valuing Nature: lessons learned
and research direction. Ecological Economics 46 (3), 493-510.
Tzoulas, K., Korpela, K., Venn, S., Yli-Pekonen, V., Kaierczak, A., Niemela, J., James, P., 2007. Promoting
ecosystem and human health in urban areas using Green Infrastructure: A literature review. Landscape and
urban planning 81 (3), 167-178.
Valero, F., Arbs, R., 2010. Desalination of brackish river water using Electrodialysis Reversal (EDR) Control of
the THMs formation in the Barcelona (NE Spain) area. Desalination 253 (1-3), 170-174.
Villanueva-Belmonte, C., 2003. Subproductes de la desinfecci de lAigua Potable i Cncer de Bufeta Urinria.
Doctoral Dissertation. Universitat Autonoma de Barcelona. Bellaterra, Spain.
Webb B.W., Hannah D.M., Moore R.D., Brown L.E., Nobilis F., 2008. Recent advances in stream and river
temperature research. Hydrological Process 22, 90218.
Worster, D., 1985. Rivers of Empire. Water, Aridity, and the Growth of the American West. Oxford University
Press. New York.
Wunder, S., 2007. The efficiency of payments for environmental services in tropical conservation. Conservation
Biology 21 (1), 48-58.
Young, R.F., 2010. Managing municipal green space for ecosystem services. Urban Forestry & Urban Greening 9
(4), 313-321.

114.
CHAPTER 4. MANAGING ECOSYSTEM SERVICES TO MEET STREAM
TEMPERATURE OBJECTIVES IN THE LLOBREGAT RIVER, SPAIN

Abstract
Ecosystem services have the potential to be incorporated into decision making more often if research

were to focus on the demand for these services rather than the supply. This demand-oriented approach

could catalyze changes in ecosystem management by directing more attention to the needs of resource

users. A demand-oriented approach also implies inverting the sequence in which the linkages between

environmental and economic systems are studied. Instead of tagging economic values to well known

biological processes, a demand approach first examines the economic, decision making and technological

context of the end-user. This chapter provides an example of how this research approach for ecosystems

services could unfold. In the Llobregat River in northeastern Spain, higher stream temperatures require

water treatment managers to switch on costly water treatment equipment during warm months. This

creates an opportunity to align the economic interests of downstream water users with the environmental

goals of river managers. A restored riparian forest or an increase in stream flow could reduce the need for

this expensive equipment by reducing stream temperatures below critical thresholds. I used the Stream

Network Temperature Model (SNTEMP) to test the impact of increasing shading and discharge on stream

temperature at the intake of the drinking water treatment plant. The value of the stream temperature

ecosystem services provided by existing forests is 79,000 per year, while the restoration of riparian

forests could generate economic savings for water treatment managers in the range of 57,000-156,000

per year, depending on the extent of the forest restoration. Stream restoration at higher elevations would

yield greater benefits than restoration in the lower reaches. Moderate increases in stream discharge (25%)

could generate savings of 40,000 per year.

Keywords: Ecosystem services, stream temperature, SNTEMP, riparian restoration, drinking water
treatment

115.
4.1 Introduction
The extensive research on ecosystem services has not generated a commensurate transformation

in how ecosystems are managed in practice. If ideas about ecosystem services are to make a unique

contribution to environmental management, the discussion must move from theoretical frameworks to

practical applications (Cowling et al. 2008, Daily and Matson 2008, Daily et al. 2009, Muradian et al

2010). I argue that ideas about ecosystem services would be more widely adopted if researchers were to

focus on the demand for ecosystem services rather than their supply. This demand-oriented approach

implies inverting the sequence in which linkages between ecological and economic systems are frequently

studied. The approach would begin with the end users, their policy objectives, or decision making and

technological context. This starting point is more likely to uncover opportunities for the management of

ecosystem services and the implementation of these ideas by resource managers.

At its core, the field of ecosystem services studies the linkages between economic and ecological

systems (MA 2005, NRC 2005). However this inquiry often begins with the ecological or biophysical

(Kremen and Ostfeld 2005, Chan et al. 2006, Naidoo et al. 2008). Researchers will frequently define the

ecological site of interest a protected area, a forest ecosystem, a watershed, or a river corridor and then

identify the range of services obtained from these areas, such as flood protection, recreational values or

spiritual values. Questions concerning economic valuation are left to the end, so by the time economists

commence the valuation exercise, the ecosystem scale and end-users have either been pre-determined or

are only vaguely identified as society. This sequence can lead to poorly targeted research, sub-optimal

management or controversial valuation estimates (Sangenberg and Settele 2010). In this chapter I

contend that ideas about ecosystem services would be more widely adopted if we studied the management

objectives that ecosystem services could help us achieve. However to do so, we must understand the

context in which decisions are made and how ecosystems may generate value. This decision making

context may then guide our inquiry, and help us identify services provided by ecosystem structures and

116.
functions. This demand-oriented approach directs our attention to the needs of resource users first, and

then seeks to show that the interests of resource users and managers may be aligned.

This chapter provides an example of how this research approach for ecosystem services could

unfold. Water users at the Aiges Ter-Llobregat (ATLL) water treatment facility in Abrera, Spain, treat

surface water from the Llobregat River to supply the Barcelona metropolitan region. Treatment managers

rely on stream temperature to guide major operation decisions. As stream temperatures rise in the spring

and summer, water managers progressively turn on electrodialysis reversal (EDR) treatment modules

(Valero and Arbs 2010). EDR modules must be added because warmer water accelerates the formation

of harmful disinfection by-products during the treatment process (Villanueva-Belmonte 2003, Sorlini and

Collivignarelli 2005, ATLL 2008). The additional EDR modules ensure that the output water quality will

comply with drinking water legislation during warmer months (ATLL 2008, Valero and Arbs 2010). As

a result of these treatment protocols, a reduction in stream temperatures would reduce the number of days

in which the expensive EDR modules would be needed.

Direct solar radiation and air temperature are the primary determinants of stream temperature,

although the river also captures heat generated from friction with the stream bed, and long wave radiation

emitted by surrounding topography and vegetation. Streams are generally coolest at their headwaters and

then temperatures rise as water moves downstream, rapidly at first, and then more slowly at lower

elevations. Urbanization, deforestation and water withdrawals also contribute to thermal heating in

streams (Webb et al. 2008). In contrast, hypolimnetic water releases from dams usually cool streams.

River managers can alter stream temperature by restoring riparian forests that block direct thermal

radiation, by increasing stream discharge, or by restoring stream meanders and surface-subsurface

interactions.

Thermal heating in rivers and streams can disrupt important ecological processes (Webb et al.

2008). Warmer waters may limit fish reproduction, accelerate ecological metabolism, and increase the

vulnerability of aquatic life to disease (Acua and Tockner 2009). Temperature also plays an important
117.
role in determining the availability of dissolved oxygen for freshwater organisms. Low levels of

dissolved oxygen during summer months are associated with major fish kills (Graczyk and Sonzogni

1991).

Modeling of stream temperature has been widely used to predict the potential effects of

management options (Bartholow 1991, Chen et al. 1998, Bartholow 2000 and 2000b; Watanabe et al.

2005). Aquatic ecologists originally developed stream temperature models to study habitat suitability for

fish, particularly in the Pacific Northwest of the United States where Federal and State agencies have

regulated maximum stream temperatures to protect endangered salmon (Caissie 2006, Webb et al. 2008).

Most stream temperature studies use deterministic models that calculate the total heat fluxes in the river

system. Others have used models based on regression techniques in which air temperature is the primary

input parameter; or stochastic modeling methods that separate annual temperature cycles from short-term

components (Caissie 2006). Temperature models that rely on calculations of the streams heat budget

allow model users to evaluate the effects of altered conditions affecting solar radiation.

Riparian forests moderate water temperatures by protecting the river from direct sunlight

(Bartholow 1989, Larson and Larson 1996, Bescheta 1997). Yet the precise impact of stream shading on

stream temperature varies with baseline conditions, vegetation height, latitude, and stream discharge.

These contextual differences make it difficult to compare shadings impact on stream temperature even

when values are translated into temperature reductions per kilometer.

Nevertheless, a few examples are illustrative. In the Cache la Poudre River, Colorado it has been

estimated that the doubling of shading from 13% to 23% would reduce maximum temperatures in the

summer by approximately 1.25C over a 32 km reach (0.04C/km) (Bartholow 1991). In contrast Seedang

et al (2008) estimated that shading decreased maximum temperatures by less than 1C (0.01C/km) in a

92 km reach in the upper main stem of the Willamette River, Oregon. The authors hypothesize that the

voluminous discharge and wide stream width prevented shading from altering stream temperature.

118.
Increasing stream discharge is another management option that may mitigate thermal heating.

Again in the Cache la Poudre River, Colorado, it was estimated that increasing discharge by 3 m3/s

(300%) would allow the reach to comply with temperature requirements year round (Bartholow 1991). In

contrast, Seedang et al (2008) found increasing discharge in the upper main stem of the Willamette River

had limited impact on temperatures.

In this study I used a deterministic stream temperature model to explore how ecosystem services

associated with river shading and discharge might reduce stream temperatures at the intake of a drinking

water facility in Abrera, Spain. This allowed me to avoid measuring all ecosystem services provided by

the Llobregat River, such as nutrient cycling, water provision, or flood protection etc, and instead target

those services for which there is a known demand from specific water users.

4.2 Study Area


The Llobregat River flows 170 km in a southward direction from its headwaters in the Pyrenees

Mountains to the Mediterranean Sea (Fig. 4.1). The upper segments of the Llobregat watershed receive

1000 mm of precipitation per year while the mid section of the watershed is considerably drier with only

400 mm per year (Mujeriego 2006). The Llobregat River has an annual discharge of 660 hm3 although its

flow regime is highly variable. The Llobregat has one major tributary, the Cardener River, which is

slightly smaller in size when the two rivers join at mid-watershed.

The Llobregat River provides the Barcelona Metropolitan Region (pop. 3.5 million) with 45% of

its drinking water supply (Mujeriego 2006). Approximately one quarter of the regions water is treated at

the ATLL treatment facility in Abrera (max. capacity = 4 m3/s). The average flow at the ATLL facility is

8 m3/s. Two major dams hold back the Llobregat and the Cardener Rivers before they leave the Pyrenees

Mountains. The size of the hypolimnetic releases from the dams is negotiated periodically between the

hydroelectric dam operators, water suppliers and the regional water agency of Catalonia (ACA).

119.
The study reach consists of 57.6 km between the stream gage at Balsareny, downstream of the

lowermost dam, and the ATLL water treatment facility in Abrera. The entire river network could not be

modeled because of the presence of the dams. The Llobregat River travels 27.8 km between Balsareny

and the confluence with the Cardener. The river then travels another 29.8 km to the ATLL water

treatment facility, located 29.0 km upstream from the rivers mouth in the Mediterranean.

Figure 4.1 The study reach is 57.6 km between the town of Balsareny and the ATLL drinking water
treatment facility in Abrera. Meteorological data was collected from a weather station at Pont de
Vilomara.

4.3 Methods
SNTEMP Model

The Stream Network Temperature Model (SNTEMP) is a one-dimensional heat transport model

for stream networks that predicts mean daily water temperatures based on heat flux equations (Theurer et

120.
al. 1984, Bartholow 2000). The mechanistic model was originally developed by the U.S. Fish and

Wildlife Service (Theurer et al. 1984) and is now distributed by the United States Geological Survey

(USGS). USGS has generated training materials for SNTEMP which has facilitated its widespread use

(Hendrick and Monahan 2003; Norton and Bradford 2009). SNTEMP is a well characterized and tested

model that can evaluate changes in both stream shading and discharge, with only moderate data

requirements. The model calculates the heat gained or lost from a parcel of water as it passes through the

various nodes in a stream network (Appendix E). The model simulates heat flux processes of convection,

conduction, evaporation, direct solar radiation (short wave), atmospheric radiation, radiation from riparian

vegetation (long wave), and back radiation released by the water (Bartholow 2000) (Appendix F). The

model requires data on stream hydrology, stream geometry, meteorology and shading conditions (Table

4.1) (Appendix G). Given these inputs, the model predicts the stream temperature at the end of the

segment (Appendix H). The stream network model was constructed using 22 nodes, each of which marks

either a change in stream temperature, discharge, geographic or shading conditions or thermal mixing

(Fig. 4.2). On average, SNTEMP models have 30 nodes (Bartholow 2000).

Table 4.1 Input data requirements for the SNTEMP Model


Stream Hydrology Stream Geometry Meteorology Shading Conditions
Segment Inflow Node Latitude Mean Daily Air Temp Topographic Shade
Temperature Inflow Node Elevation Relative Humidity Vegetation Shade
Segment Outflow Width A and B Terms Wind Speed
Accretion Temperature Mannings N Solar Radiation
Ground Temperature
Thermal Gradient

The Llobregat River SNTEMP Model


I obtained meteorological data from the Catalan Meteorological Service that operates a weather

station located at the center of the stream segment (km 37) in the town of el Pont de Vilomara (406310,

121.
4617994 UTM) (Fig 4.1). Hydrologic and stream temperature data were obtained from the Catalan Water

Agency (ACA). A complete data set for all nodes was available for the 2009 calendar year.

Figure 4.2 A schematic representation of the modeled stream network between Balsareny and the ATLL
drinking water treatment facility in Abrera, Spain. Data on stream discharge is required for Headwaters
(H), Discharge Verification (Q), Point Discharges (P), and End (E) nodes. Temperature data is input at
Headwaters (H), Tributaries (T), Branch (B), and Point Discharges (P) nodes.

I measured vegetative and topographic shade on 8 and 9 May 2011 following the field work

guidelines outlined in Bartholow (1989) (Appendix I). I verified the location of stream gages, diversions,

returns and point sources, and identified sites where the properties of the stream changed (C nodes). I

also evaluated if the restoration of riparian forests was viable given current soil quality or topographic

conditions. I annotated the location of sites where extreme slopes (cliffs), or the absence of soils (large

stones and boulders) would make any proposal to reforest unrealistic. Forest coverage estimates

122.
collected in the field were compared with satellite imagery available from Google Earth. In stream

segments where field measurements were not taken I relied on satellite imagery.

Shading Scenarios

I divided the Llobregat into multiple segments according to topographic and forest conditions. I

defined a stream segment as restorable if it had less than 30% forest cover along its bank and did not

have an obvious impediment to forest restoration such as poor soil quality, or steep topography.

I tested five shading scenarios. The first three scenarios increased shade only in the subset of

river segments defined as restorable (<30% cover and favorable soil quality and topography). Shading

was increased from its existing condition (<30%) to 40%, 60% and 80% forest cover. The next shading

scenario increased shade for the entire stream segment to 100% coverage. This scenario serves as an

upper boundary estimate of shadings potential impact. In the final scenario, I removed all existing

riparian forest to estimate the contribution of current shade in moderating stream temperatures.

Discharge Scenarios

I tested five discharge scenarios that increased stream flow by 5%, 25%, 50%, and 100% of

historic monthly means. Monthly mean discharge values were computed from a historic data set of six

years (2004-2009).

Model Calibration

The calibration of predicted to observed temperatures was accomplished by adjusting calibration

parameters built into the model and following recommended procedures (Bartholow 1991). In this case I

only adjusted the coefficient for air temperature across seasons. During winter months, from November to

February, the air temperature calibration coefficient was 0.25. During the peak summer months of June to

September it was 0.9. In the remaining months the air temperature calibration coefficient was 0.5
123.
(Appendix J). It was important to ensure that the models predictions approximated observed outcomes.

Although for my purposes, it was even more important that the model help me accurately estimate the

impact of the various management scenarios on stream temperature.

Value of Ecosystem Services

Managers at the ATLL treatment plant begin to add EDR modules in March of each year when

stream temperatures surpass 12C. At peak summer temperatures all nine EDR modules are in operation.

Since one EDR module is in operation year round, there are eight additional modules that may be added

in warm months. This generates eight temperature thresholds in which the costly EDR treatment modules

are added. Although the precise protocol for adding EDR modules has not been formalized by ATLL, a

linear relation to temperature provides a reasonable decision rule (personal communication, Valero 2010).

In this study, I assume an EDR module is added for every 2C rise in temperature beyond 12C (14, 16,

18, etc...). Each module consumes approximately 12,000 kWh per day for a cost of 1,000/day (personal

communication, Barcel 2010). Value is generated when ecosystem processes delay the exceedance of

one of the eight critical temperature thresholds by at least one day and therefore allow water treatment

managers to avoid turning on an additional module.

Riparian Reforestation Costs

I collected information on riparian restoration projects in order to estimate the costs associated

with the increasing stream shade along the Llobregat Rivers and conduct a cost-benefit analysis of the

restoration projects over a 20 year period. ACA has financed projects that remove invasive species of

cane (Arundo donax) and replaced them with native species of Willow (Salix alba), Ash (Fraxinus

agustifolia) and Poplar (Populus Alba). These projects were executed along the Llobregat and Cardener
124.
Rivers, and therefore offer a useful precedent to help gage future restoration costs (Consell Comarcal del

Bages and Phragmites SL 2010 and 2011). Given the initial investment, I calculated the net present value

of the project over a 20 year period with a discount rate of 4%. Since we cannot expect that the full

benefits associated with shading will materialize until the trees reach maturity, I estimated annual savings

that were proportional to the growth rate of riparian trees (Wanatabe et al 2005).

Robustness Tests

While the SNTEMP model has already been validated (Bartholow 1991, Bartholow 2000), the

model does not provide users with estimates of uncertainty or error. Therefore I tested the robustness of

my results to assumptions concerning estimated parameters (Mannings N) and their sensitivity to

measurement error (relative humidity, stream width, stream height). In the stream shading scenarios, I

modified relative humidity (+/- 10%), stream width (+/- 20%), vegetation height (+/- 25%) and tested

Mannings N at 0.045 and 0.025. For the stream discharge scenarios, I re-calculated the results with

changes in relative humidity (+/- 10%), stream width (+/- 20%) and vegetation height (+/- 25%). The

percent +/- variation was chosen to be conservative, or approximate the maximum potential deviance, and

therefore serve as an upper and lower boundary of expected error. To test each parameter I ran the model

20 times, testing for variation by a given percentage under two new baseline scenarios (+/-), and then

running every vegetation and discharge scenario separately (9 scenarios total, each run twice +/-) and

comparing the new results to the respective baseline scenarios (2). I also tested for robustness to

assumptions pertaining to model calibration and to the location of EDR thresholds.

4.4 Results
SNTEMP produced satisfactory water temperature predictions during the spring, summer and fall

of 2009 (R2=0.93, N=255, Pearson correlation=0.94) (Fig 4.3). The calibrated model predicted stream

temperatures with a mean error of 0.53C, overestimating observed temperature. Forty percent of the

125.
models predictions fell within 1C of the observed value. The maximum error was 4.6C. To remove the

effects of this systematic overestimate in my valuation of ecosystem services, I regressed modeled versus

observed temperature and adjusted model output accordingly (Fig 4.4).

Figure 4.3 Observed and predicted temperature values for the Llobregat River during the warmer months
of 2009 in which EDR modules are in operation at the drinking water facility.

126.
Figure 4.4 Observed and modeled temperatures for all dates in 2009. EDR water treatment modules are
only used when temperatures surpass 12C. At these warmer temperatures the SNTEMP model better
predicts observed temperature. (R2=0.933) T= 0.8068(observed temp) + 4.335. The bold line marks the
1:1 relationship.

Robustness of Temperature Estimates

When testing the sensitivity of temperature predictions to changes in humidity, stream width, and

vegetation height, mean annual temperature values changed modestly, and always within 1C. As

expected, the predicted values were always bounded within the higher and lower estimates (+/-) (Fig 4.5

& 4.6). For both the vegetation and discharge scenarios, temperature results were most sensitive to

changes in humidity.

127.
Figure 4.5 The sensitivity of temperature estimates (mean annual) in the different vegetation scenarios
when looking at changes in humidity (+/- 10%), stream width (+/-20%) and vegetation height (+/-20%).
In all cases, the results were bracketed by the higher and lower estimates. The variation was always
slightly below 1C.

128.
Figure 4.6 The sensitivity of temperature estimates in the different discharge scenarios when looking at
changes in humidity (+/- 10%), stream width (+/-20%) and vegetation height (+/-20%).

Effect of shading and discharge on temperature

An increase in shade over the Llobregat River showed reduced temperatures throughout the

stream segment, while the scenario in which all vegetation was removed increased water temperature.

Shade produced a stronger effect in warmer months (Fig 4.7). The most modest shading scenario

(restorable areas to 40%) reduced mean annual stream temperature at the intake of the ATLL water

treatment facility by 0.34C (0.006C/km), while a simulated forest coverage of 100% on all segments

showed a mean annual reduction of 1.84 C (0.03C/km) (Appendix K). The removal of existing riparian

vegetation increased stream temperature at the ATLL treatment plant by over 0.5C between May and

October.

Increasing stream discharge also reduced stream temperature. A 5% increase in discharge had a

very small effect on stream temperature (mean reduction = 0.048C), whereas a 25% increase in

discharge reduced temperature 0.2C during warm months (Fig 4.8). An increase in discharge by 50%
129.
reduced stream temperatures by 0.4C during warm months. The doubling of discharge (100%) had a

greater effect in the spring and fall, when stream discharge is largest. I hypothesize that the doubling of

discharge in the summer made little difference because the summer base flows are already small.

Figure 4.7 Mean monthly temperature changes simulated by SNTEMP in the five shading scenarios: (1)
existing vegetation coverage removed, (2) riparian forest in restorable areas increased to 40% coverage,
(3) riparian forest in restorable areas increased to 60% coverage, (4) riparian forest in restorable areas
increased to 80% coverage, (5) 100% coverage in the entire stream reach.

130.
Figure 4.8 Mean monthly temperature changes simulated by SNTEMP in the four discharge scenarios:
(1) increasing discharge 5% of monthly mean, (2) 25% of monthly mean, (3) 50% of monthly mean, and
(4) 100% of monthly mean.

Longitudinal Temperature Profile


The simulated reductions in stream temperature could prevent the Llobregat River from crossing

critical temperature thresholds, and therefore save water managers from adding EDR modules on

particular days. Each day in which water managers avoid adding an additional EDR module would save

them 1000. The temperature profile on a particular day allows us to visualize the induced changes in

stream temperature projected by each management scenario. For example, on 26 August 2009, one of the

warmest days of the year, the stream temperature rose rapidly in the first half of this segment and then

more gradually between Castellbell (km 24) and the ATLL treatment plant (Fig 4.9). On this day modest

changes in riparian shading and discharge were insufficient to maintain stream temperatures below the

critical threshold of 26C. Only the most aggressive measures, such as the doubling of discharge or

restoring 100% coverage would have kept temperatures below the critical threshold. In contrast,

131.
removing existing vegetation would have increased stream temperature to 27.72C at the intake of the

ATLL water treatment plant.

Llobregat River Temperature Profile


26 August 2009
28

26

Stream Temperature (C)


No Coverage
24
Baseline

Restorable Areas to 40%


22
100% Discharge
20 100% Coverage

18

16
60 50 40 30 20 10 0
Distance upstream (Km)

Figure 4.9 The temperature profiles under different simulated management scenarios on 26 August 2009.
The horizontal lines represent temperature thresholds. On this day, the waters of the Cardener River had a
slight cooling effect that can be appreciated by a dip in stream temperature at the river confluence (km
29).

On other days, modest management actions were sufficient to prevent the stream temperature from

exceeding critical thresholds. For example on 25 May 2009 increasing forest cover to 40% in restorable

areas would have reduced stream temperature by 0.45C and maintained stream temperature below the

20C threshold at the end of the segment (Fig 4.10). For this date, then, water managers would have

operated without an additional EDR module, saving them 1000 in this management scenario.

132.
Llobregat River Temperature Profile
25 May 2009
22

21

20

Stream Temperature (C)


19

18 No Coverage
Baseline
17 Restorable to 40%
50% Discharge
16
100% Coverage
15

14

13
60 50 40 30 20 10 0
Distance Upstream (km)

Figure 4.10 The temperature profiles under different simulated management scenarios on 25 May 2009.
The dashed lines represent the temperature thresholds. The horizontal lines represent temperature
thresholds.

In both longitudinal profiles, temperatures rose faster in the first half of the stream segment,

between the stream headwaters and the confluence with the Cardener (29.8 km). This rapid thermal

heating in the upper parts of the watershed has been documented elsewhere in the literature (Chan et al.

1998).

Avoided Threshold Crossings

As expected, large increases in shade and discharge are predicted to prevent more threshold

crossings (Fig 4.11 and 4.12). The shading scenarios would prevent between 57 and 283 threshold

crossings, while the discharge scenarios would prevent between 7 and 120 threshold crossings.

133.
Figure 4.11 Avoided threshold crossings associated with riparian restoration scenarios measured in days.
Positive values are days in which the water treatment plant could avoid operating an additional EDR
module. Negative values are days in which the treatment plant would need to operate an additional EDR
module, which is only the case in the scenario when existing revegetation is hypothetically removed.

Figure 4.12 Avoided temperature threshold crossings, measured in days, associated with a percent
increase in stream discharge: 5%, 25%, 50%, and 100%.

134.
Ecosystem services associated with stream shading

By adding the number of days in which threshold crossings are prevented, the five shading

scenarios could generate savings in the range of 57,000 to 283,000 per year (Fig 4.13). Our model also

allowed us to quantify the value of the services provided by the existing vegetation, which is estimated by

the scenario in which vegetation is removed. Removing the existing vegetation would require treatment

managers to switch on EDR modules an additional 79 days during the year, for an annual cost of 79,000.

Figure 4.13 Value of ecosystem service for the ATLL water treatment plant of various riparian
restoration options.

Ecosystem services associated with increased discharge

Increasing stream discharge by 5% had a nearly negligible impact on stream temperature,

preventing threshold crossings on only 7 days for a savings of 7,000 annually. In contrast a 25%

increase in discharge would prevent 40 threshold crossings per year, saving the drinking water facility

40,000 annually. Increasing discharge by 50% would prevent threshold crossings on 75 days, saving

water treatment managers 75,000 annually, while the doubling of discharge (100%) would prevent

threshold crossings on 120 days, saving water treatment managers 120,000 annually (Fig 4.14).

135.
Figure 4.14 Value of ecosystem service for the ATLL water treatment plant for increases in stream
discharge (% of monthy mean).

Robustness of the Value of Ecosystem Services

I also found that the value of the ecosystem services were highly robust to changes in modeling

assumptions or possible measurement error (Fig 4.15 and Fig 4.16). The significant changes in estimated

or measured parameters show relatively modest changes in the results of the economic valuation. The

sign and direction of the values remained similar in all the scenarios tested. However unlike the

sensitivity tests on temperature values (Fig 4.5 and 4.6), the economic valuation estimates were not

evenly bounded because economic value was only generated when thresholds were crossed.

136.
Figure 4.15 The upper and lower boundary results in the vegetative scenarios when testing for possible
error in relative humidity (+/-10%), , stream width (+/- 20%), vegetation height (+/-25%), and Mannings
N (0.0025 & 0.0035).

Figure 4.16 The upper and lower boundary results in the discharge scenarios when testing for possible
error in relative humidity (+/-10%), stream width (+/-20%) and vegetation height (+/-25%).

137.
I also tested if the results were sensitive to model calibration (Fig 4.17). The uncalibrated model

showed values very similar to the final results. The insensitivity of the results to model calibration is

explained by how the economic values are calculated. Value is created for water treatment managers

when thresholds are crossed, therefore even temperature models that over- or under-estimates temperature

values (uncalibrated) will still predict temperature reductions, and therefore threshold crossings. The

similarity of the results between the calibrated and uncalibrated models gives me considerable confidence

that my final results are robust to changes in model calibration.

Figure 4.17 Comparison of results between the calibrated and uncalibrated models.

Finally, I tested if the results were sensitive to the placement of thresholds at particular

temperature values. Recall that the temperature thresholds for adding an additional EDR model were

uniformily distributed every 2C after 12C (12C, 14C, 16C, 18C, 20C, 22C, 24C, 26C). I tested

how the results might change if these thresholds were 1C lower (11C, 13C, 15C, 17C, 19C, 21C,

23C, 25C) as well as thresholds that were randomly generated (12.8C, 13.9C, 15.7C, 18.6C, 21.8C,

22.4C, 23.7C, 24.7C). The results are not hugely sensitive to these changes (Fig 4.18), suggesting that

138.
even if water managers modify the temperature thresholds for turning on EDR modules, the value of the

ecosystem services are unlikely to change significantly.

Figure 4.18 The results were largely insensitive to the placement of the EDR thresholds. When reducing
the thresholds uniformly by 1C, the results changed only slightly. Nor did the results change
substantially when the temperature thresholds were selected randomly.

Forest Restoration Projects

Past restoration projects along the Llobregat and Cardener Rivers suggest that restoration costs

are approximately 120,300 per kilometer (both sides), or 6.02/m2. At this rate, the restoration cost for

each scenario is 1,082,700; 1,624,050; 2,165,400 and 3,464,640 respectively. These upfront

investment costs would be partially recovered through savings in reduced water treatment costs. In each

of the restoration scenarios, at least one third of the initial investment is recovered within a 20 year period

(Table 4.2). The most ambitious restoration scenario recovered up to 46% of the investment. Each of

these projects would have a positive net benefit in 20 years if restoration costs could be lowered to

20,000 per kilometer, or 2.0/m2 (Appendix O).

139.
Table 4.2 Summary of restoration costs and % investment recovered after 20 years.

Restoration Scenarios Segment (Km) Cost () % Investment Recovered

Recoverable Areas to 40% 18 1,082,700 33%


Recoverable Areas to 60% 27 1,624,050 41%

Recoverable Areas to 80% 36 2,165,400 46%

All areas 100% 57.6 3,464,640 46%

Increases in Discharge

Dam releases in two reservoirs (Sant Pon and Baells) in the Llobregat watershed could increase

water flow at a low cost. Moreover, in contrast to riparian restoration, the reductions in stream

temperature associated with dam releases would materialize immediately. The critical tradeoff is the loss

of water storage. When water supplies are low, the dam releases may not be a viable option. But if

temperatures are rising and water supplies are abundant, dam releases may be an effective measure to

postpone the use of expensive EDR water treatment.

4.5 Discussion
The existing stream vegetation along the Llobregat River already provides valuable ecosystem

services for the ATLL water treatment plant, and by extension, for the residents of Barcelona, by

maintaining the Llobregat River cool and avoiding the use of EDR modules on approximately 79 days per

year, saving 79,000 per year in water treatment costs. Restoring the Llobregat River with more riparian

trees or increasing discharge could provide even greater benefits.

The least ambitious shading option would generate services worth 57,000 per year, while the

most ambitious scenario, the restoration of vegetation along the entire stream segment (100% coverage),

could save water managers 283,000 per year. This last scenario serves as an upper boundary estimate

140.
and also shows that ecosystem processes have the potential to generate substantial value for downstream

users.

Increases in stream discharge also have the potential to generate additional savings for water

treatment managers. The timing of discharge increases is particularly important (Fig 4.8). Greater flows

are most likely to reduce stream temperature the months of March, April, May, August, September and

October.

These results have implications for water managers in Catalonia as they seek to comply with the

Water Framework Directive (WFD) that requires that all water bodies obtain good ecological status

(Directive 2000). In an effort to comply with the WFD, the Catalan Water Agency has drafted a plan of

measures, published in 2010, that budgeted 8.67 million for riparian restoration along the Llobregat

River (ACA 2010). This research shows that some of these costs can be recovered. This is useful

information as policy makers must decide which measures to execute and which measures might be

postponed because of their financial cost. Recall that governments may be excempt from the European

Unions requirement to restore water bodies to good ecological status if they can justify that the costs

associated with ecological restoration are disproportionately large. However these economic assessments

frequently only focus on the costs of the restoration measure and not the associated environmental and

economic benefits, as presented here. Therefore this research helps quantify some of the benefits

associated with ACAs program of measures. At the same time, the WFD seeks to invert our economic

logic by putting the objective of good ecological status at the center of policymaking. Following this

logic, the restoration benefits quantified in this paper are in addition to our policy objective of good

ecological status rather than a justification for them.

The model does not include the inputs from all waste water treatment plants along this river

segment. However when two waste water treatment plants were included in the lower reach they had a

negligible impact on downstream temperature. The model also does not consider the effect of successive

river impoundments which hold back the water on its journey between Balsareny and Abrera.
141.
The longitudinal profiles suggest that the location of riparian shading matters. In this case, the

increased shading between Balsareny and Castellbell would be especially important since the river gains

the most heat in this segment. As a result, it is here where management is likely to have the most impact

(Appendix L). Other watersheds have also shown a pattern of streams heating up faster in the upper parts

of the watershed (Chen et al. 1998).

None of the restoration scenarios could offer projects with a positive net present value after 20

years because of the high initial investment and low initial returns that do not become more substantial

until the trees reach full maturity. If the restoration projects were to plant older trees they would provide

more benefits earlier, and the projects would have higher returns.

Increasing stream flow through strategic dam releases would allow water managers to realize

benefits within a shorter time frame. These releases would need to be negotiated with the Catalan Water

Agency and the electrical company that captures energy from the hydroelectric dams. The institutional

structures already exist to allow this negotiation to develop. Additional releases would be most realistic

when reservoirs are near full capacity and air temperatures are rising rapidly. For example, if reservoirs

were full in late May due to spring rains, and temperatures were to rise abruptly, the additional discharge

could help maintain cooler stream temperatures, and prevent the ATLL treatment plant from turning on

additional EDR modules for several days or weeks.

This chapter has applied a deterministic model of biophysical processes to simulate alternative

management options in the watershed. And yet the biophysical model alone could not have generated the

reliable valuation estimates of the ecosystem services related to stream temperature. The technological

and decision making conditions downstream dictated the terms under which valuation estimates could be

made. This implies that different technologies or alternative management protocols would alter the value

associated with ecosystem services that moderate stream temperature. From the perspective of the water

treatment managers, the ecosystem service was created by the new technology and the treatment protocol

that centered decision making on stream temperature values.


142.
I have calculated a value for the ecosystem services by focusing on how these services could help

treatment managers meet specific water quality targets. Focusing on how ecosystem services could

contribute to meeting policy objectives allowed me to avoid thorny questions concerning non-use

valuation. Of course, streams provide both use and non-use values. However, non-use values can be

difficult to quantify. And even when these controversial methods are accepted, there are still concerns that

valuation exercises are inappropriate or misguided (Venkatachalam 2004, McCauley 2006, Gmez-

Baggethun et al. 2010, Kosoy and Corbera 2010, Sangenberg and Settele 2010). To avoid some of these

controversial issues associated with non-use valuation, in this chapter I did not seek to value the

ecosystem structures in their entirety (use and non-use), but merely quantified the use-values generated

for a specific downstream users. Thus the results presented here offer a lower boundary estimate of the

value provided by riparian ecosystems.

Managing ecosystem services in order to meet specific objectives has several advantages. It

simplifies the valuation process, engages decision makers, and facilitates the implementation of ideas

about ecosystem services in practice. This approach may not be viable for all ecosystem services studied,

especially when non-use values are involved. However it does highlight that new opportunities to

manage and restore ecosystem services may arise if we scrutinize the decision making context of those

who depend on natures services. By focusing on the end users, we will be more likely to harness the

benefits that ecosystems provide.

143.
4.6 Acknowledgements
This research has been funded by the Water Technology Center (CETaqua) and the Catalan Water

Agency (ACA) in a collaborative agreement with the Catalan Water Research Institute (ICRA), Aiges-

Ter Llobregat (ATLL) and Aiges de Barcelona (AGBAR). Results were obtained with data from the

Meteorological Service of Catalonia (Meteocat), Aiges Ter Llobregat (ATLL), and ACA. ACA also

provided the HEC-RAS model to calculate the width A and B constants, with valuable assistance from

Rosana Aguilera at ICRA and Ramn Batalla from the University of Lleida. I thank Maria Jou from

Mina-Sorea for providing data on discharge and temperature from the wastewater treatment plants. John

Risley and John Bartholow from the United States Geological Survey (USGS) provided valuable

assistance developing the Stream Network Model. I thank Francesc Solana for meteorological data used

to produce preliminary results for Stream Segment Model. ric Esclassans and Hctor Oliva loaned me

equipment to conduct the field work.

144.
4.7 References Chapter 4

ACA. 2010. Programa de measures del Pla de gesti del districte de conca fluvial de Catalunya. Annex II. Llistat
dactuacions. Generalitat de Catalunya. Departament de Medi Ambient i Habitatge. Barcelona, Spain.
Acua, V. and K. Tockner. 2009. Surface-subsurface water exchange rates along alluvial river reaches control the
thermal patterns in an Alpine river network. Freshwater Biology 54: 306-320. Doi:10/1111/j.1365-
2427.2008.02109.x
Acua, V., A. Wolf, U. Uehlinger and K. Tockner. 2008. Temperature dependence of stream benthic respiration in
an Alpine river network with relevance to global warming. Freshwater Biology 53: 2076-2088.
ATLL. 2008. Estudi per a loptimitzaci econmica-sanitria del funcionament conjunct del tractament
convencional I la instal.laci delectrodilisis reversible a lETAP del Llobregat (T.M. Abrera).
Planificaci del servei de producci daigua per al consum hum. Aiges Ter Llobregat. 17 September
2008.
Bartholow, J.M. 2004. SSTEMP for Windows: The Stream Segment Temperature Model (Version 2). [Available
online at http://www.fort.usgs.gov/Products/Publications/10016/10016.pdf]
Bartholow, J.M. 2003. Modeling uncertainty: Quicksand for water temperature modeling. Hydrological Science and
Technology 19(1-4): 221-232.
Bartholow, J.M. 2000. The Stream Segment and Stream Network Temperature Models: A Self-Study Guide. US.
Geological Survey. US Department of the Interior. Version 2.0 March 2000. Open File Report 99-112. US
Geological Survey computer model and documentation. [Available online at
http://www.fort.usgs.gov/products/software/SNTEMP/]
Bartholow, J.M. 2000b. Estimating cumulative effects of clearcutting on stream temperatures. Rivers 7(4): 284-297.
Bartholow, J.M. 1991. A Modeling Assessment of the Thermal Regime for an Urban Sport Fishery. Environmental
Management. 15(6): 833-845.
Bartholow, J.M. 1989. Stream Temperature Investigations: field and analytic methods. U.S. Fish and Wildlife
Service. Instream Paper No. 13. Biological Report 89: 17. Washington D.C.
Beschta, R.L. 1997. Riparian Shade and Stream Temperature: An Alternative Perspective. Rangelands 19(2): 25-28.
Boyd, M. and D. Sturdevant. 1997. The scientific basis for Oregons Stream Temperature Standard. Common
Questions and Straight Answers. Oregon Department of Environmental Quality, Salem, Oregon.
Cassie, D. 2006. Thermal regime of rivers: a review. Freshwater Biology 51: 1389-1406.
Chan, K.M.A., M.R. Shaw, D.R. Cameron, E.C. Underwood and G.C. Daily. 2006. Conservation planning for
ecosystem services. PLoS Biology 4 (11): 2138-52.
Chen, Y.D., S.C. McCutcheon, D.J. Norton and W.L. Nutter. 1998. Stream Temperature Simulation of Forested
Riparian Areas II. Model Applications. Journal of Environmental Engineering 124(4): 316-328.
Consell Comarcal del Bages and Phragmites SL. 2011. Informe sobre leradicaci de la canya Arundo donax.
Report to ACA (internal document).
Consell Comarcal del Bages and Phragmites SL. 2010. Informe sobre leradicaci de la canya Arundo donax.
Report to ACA (internal document).
Constanz, J. 1998. Interactions between stream temperature, streamflow, and groundwater exchanges in alpine
streams. Water Resources Research 34(7): 1609-1615.
Countant, CC. 1985. Stripped Bass, Temperature, and Dissolved-Oxygen A speculative hypothesis for
environmental risk. Transactions of the American Fisheries Society 114(1): 31-61.
Daily, G.C., S. Polasky, J. Goldstein, P.M. Kareiva, H.A. Mooney, L. Pejchar, T.H. Ricketts, J.Salzman and R.
Shallenberger. 2009. Ecosystem services in decision making: Time to deliver Frontiers in Ecology and the
Environment 7 (1) (02/01): 21-8.
Diputaci de Barcelona, no date. Projecte Riu Verd a Monistrol de Montserrat. Public Signage on Llobregat River
in Monistrol de Montserrat. Photograph taken May 2010.
Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework
for Community action in the field of water policy. [Available online http://eur-
lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32000L0060:EN:NOT] Accessed 23 April 2012.

145.
Gmez-Baggethun, E., R. de Groot, P.L. Lomas and C. Montes. 2010. The history of ecosystem services in
economic theory and practice: From early notions to markets and payment schemes. Ecological Economics
69: 1209-1218.
Graczyk, D.J. and W.C. Sonzogni 1991. Reductions of dissolved-oxygen concentrations in Wisconsin streams
during summer runoff. Journal of Environmental Quality 20(2): 445-451.
Hendrick, R. and J. Monahan. 2003. An assessment of water temperatures of the Entiat River, Washington using the
Stream Network Temperature Model (SNTEMP). Report for the Entiat WRIA Planning Unit (EWPU).
[Available online at www.cascadiacd.org/files/documents/SNTEMP_FinalDraft_Sept03.pdf]
Hester, E.T. and M.N. Gooseff. 2010. Moving beyond the banks: hypohreic restoration is fundamental to restoring
ecological services and functions of streams. Environmental Science and Technology 44: 1521-1525.
Johnson, S.L. 2004. Factors influencing stream temperature in small streams: substrate effects and a shading
experiment. Canadian Journal of Fisheries and Aquatic Science 61: 913-923
Johnson, S.L. 2003. Stream temperature: scaling of observations and issues for modeling. Hydrological Processes
17: 497-499.
Kauffman, J. B., R.L. Beschta, R.L, Ottiny and D. Lytjen. 1997. An ecological perspective of riparian and stream
restoration in the western United States. Fisheries 22: 12-24.
Kosoy, N. and E. Corbera. 2010. Payments for Ecosystem Services as Commodity Fetishism. Ecological Economics
69 (6): 12281236
Kremen, C. and R.S. Ostfeld. 2005. A call to ecologists: Measuring, analyzing, and managing Ecosystem services.
Frontiers in Ecology and the Environment 3: 540-548.
Larson, L.L. and S. L. Larson. 1996. Riparian Shade and Stream Temperature: A Prespective. Rangelands 18(4):
149-152.
LeBlanc, R.T., R. D. Brown and J. E. FitzGibbon. 1997. Modeling the effects of land use change on water
temperature in unregulated urban streams. Journal of Environmental Management 49: 445-469.
Meehan, W.R. 1970. Some effects of shade cover on stream temperature in southeast Alaska. USDA Forest Service
Res. Note PNW 113. Pacific Northwest Forest and Range Experiment Station.
Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: Current states and trends. Island
Press: Washington D.C. Available online: www.maweb.org
Mujeriego, R. 2006. Abastament daigua des del Baix Llobregat nord: Diagnosi per a la millora de la qualitat.
Agncia Catalana de lAigua, Aigues Ter-Llobregat, Direcci General de Salut Pblica. Generalitat de
Catalunya. Barcelona, Spain.
Naidoo, R., A. Balmford, R. Costanza, B. Fisher, R.E. Green, B. Lehner, T.R. Malcom and T.H. Ricketts. 2008.
Global mapping of ecosystem services and conservation priorities. Proceedings of the National Academy of
Sciences 105: 9495-9500.
National Research Council. 2005. Valuing Ecosystem Services: Toward Better Environmental Decision Making.
National Academy Press: Washington, D.C.
Nelson E.G. et al. 2009. Modeling multiple ecosystem services, biodiversity conservation, commodity production,
and tradeoffs at landscape scales. Frontiers in Ecology and the Environment 1: 4-11.
Norton, G.E. and A. Bradford. 2009. Comparison of two stream temperature models and evaluation of potential
management alternatives for the Speed River, Southern Ontario. Journal of Environmental Management
90: 866-878.
Prats, J., R. Val, J. Armengol and J. Dolz. 2010. Temporal variability in the thermal regime of the lower Ebro River
(Spain) and alteration due to anthropogenic factors. Journal of Hydrology 387: 105-118.
Real Decreto 140/2003. Establishment of Spanish Drinking Water Quality Standards. 7 February 2003.
Sangenberg, J.H. and J. Settele 2010. Precisely incorrect? Monetising the value of ecosystem services. Ecological
Complexity 7: 327-337.
Seedang, S., A.G. Fernald, R.M. Adams and D.H. Landers. 2008. Economic analysis of water temperature reduction
practices in a large river floodplain: an exploratory study of the Willamette River, Oregon. River Research
and Applications 24: 941-959.
Sorlini, S. and C. Collivignarelli. 2005. Trihalomethane formation during chemical oxidation with chlorine, chlorine
dioxide and ozone of ten Italian natural waters. Desalination 176: 103-111.

146.
Sweeney, B.W., T.L. Boff, J.K. Jackson, L.A. Kaplan, J.D. Newbold, C.J. Standley, W.C. Hession and P.J.
Horowitz. 2004. Riparian deforestation, stream narrowing and loss of ecosystem services. Proceedings of
the National Academy of Sciences 101(39): 14132-14137.
Theurer, F.D., I. Lines and T. Nelson. 1985. Interactions between riparian vegetation, water temperature and
salmonid habitat in the Tucannon River. Water Resources Bulletin 21: 53-64.
Theurer, F.D., K.A. Voos and W.J. Miller. 1984. Instream water temperature model. Instream flow information
Paper 16. US Fish and Wildlife Service FWS/OBS-85/15.vp.
Thorp, J. H., J.H. Flotmersch, M.D. Delong, A.F. Casper, M.C. Thoms, F. Ballantyne, B.S. Williams, B.J. ONeill
and S. Haase. 2010. Linking ecosystem services, rehabilitation, and river hydrogeomorphology. 2010.
Bioscience 60: 67-74.
Tonolla, D., V. Acua, U. Uehlinger, T. Frank and K. Tockner. 2010. Thermal Heterogeneity in River Floodplains.
Ecosystems. DOI 10.1007/s10021-010-9350-5.
Torgersen, C.E., R.N. Faux, B.A. McIntosh, N.J. Poage and D.J. Norton. 2001. Airborne thermal sensing for water
temperature assessment in rivers and streams. Remote Sensing of Environment 76: 386-398.
Valderi-Prez R., M. Lpez-Rodrguez and J.A. Ibez-Mengual. 2001. Characterizing an electrodialysis reversal
pilot plant. Desalination 137: 199-206.
Valero, F. and R. Arbs. 2010. Desalination of brackish river water using Electrodialysis Reversal (EDR) Control of
the THMs formation in the Barcelona (NE Spain) area. Desalination 253: 170-174.
Venkatachalam, L. 2004. The contingent valuation method: a review. Environmental Impact Assessment Review
24: 89-124.
Villanueva-Belmonte, C. 2003. Subproductes de la desinfecci de lAigua Potable i Cncer de Bufeta Urinria.
Tesis Doctoral. Universitat Autonoma de Barcelona.
Watanabe, M., R.A. Adams, J. Wu, J.P. Bolte, M.M. Cox, S.L. Johnson, W.J. Liss, W.G. Boggess and J.L. Ebersole.
2005. Toward efficient riparian restoration: integrating economic, physical, and biological models. Journal
of Environmental Management 75: 93-104.
Webb B.W., D.M. Hannah, R.D. Moore, L.E. Brown and F. Nobilis. 2008. Recent advances in stream and river
temperature research. Hydrological Processes 22: 90218.
Zwieniecki, M.A. and M. Newton. 1999. Influence of streamside cover and stream features on temperature trends in
forested streams of Western Oregon. Western Journal of Applied Forestry 14: 106-113.

147.
APPENDICES

Appendix A. Osmotic Pressure, Total Dissolved Solids, & Energy Efficiency at AGBAR

Appendix B. Electrodialysis Reversal (EDR), Conductivity, & Energy Efficiency at ATLL

Appendix C. Summary of the Value of Ecosystem Services Following Technological Change

Appendix D. Additional Reflections on Ecosystem Services and Technology

Appendix E. Heat Fluxes Modeled in the Stream Network Temperature Model (SNTEMP)

Appendix F. Stream Network Model Operation

Appendix G. Input Files for the Stream Network Model (SNTEMP)

Appendix H. Output File for Data Verification Program TDATCHK

Appendix I. Field Work for Shading Estimates

Appendix J. Calibration of SNTEMP

Appendix K. Full Results for SNTEMP

Appendix L. Notes on Scale Selection for SNTEMP

Appendix M. Stream Segment Model (SSTEMP)

Appendix N. Stream Network Model from Castellbell to Abrera

Appendix O. Riparian Vegetation and Restoration Costs

148.
Appendix A. Osmotic Pressure, Total Dissolved Solids, & Energy Efficiency at
AGBAR

AGBAR spends between 1.8 and 2.2 million per year on energy to desalinate water from the Llobregat
River with reverse osmosis technology. This estimate only covers energy expenses associated with the
reverse osmosis membranes and excludes other energy use associated with ultrafiltration or water
elevation. Given these high energy expenses, a reduction in salinity concentrations in the Llobregat River
could generate considerable financial savings. In the main text I calculated these savings using the
Reverse Osmosis System Analysis (ROSA) software distributed by Dow Chemical.

Underlying these calculations are the physical laws that govern osmotic pressure. To remove dissolved
solids from a solution, the reverse osmosis process must apply a pressure that surpasses the osmotic
pressure (Elimelech and Phillip 2011). A higher concentration of the solute will increase the osmotic
pressure and therefore increase the energy needed to desalinate. Osmotic pressure () is a function of the
molar concentration of total dissolved solids (c), the gas constant (R), and temperature (T) (Equation 1).

(1) Equation for osmotic pressure: = cTR

Where:
c = the molar concentration of the solute;
R= the gas constant 0.082 (literbar)/(degmol);
T= temperature (Kelvin)

Higher concentrations of dissolved solids need more osmotic pressure for the same output water quality
(Elimelech and Phillip 2011). However it difficult to calculate the energy TDS-energy efficiency
relationship directly because RO systems have three phases, and the pressure that must be applied must
surpass the osmotic pressure in all three desalination phases. With three phases and seven membranes in
each phase, one molecule of water will have gone through 21 membranes before it exits the system. At
each phase, the rejected water that does not pass the membrane begins to accumulate salts. Hence the
minimum pressure needing application in each membrane increases as dissolved solids accumulate.

AGBAR treats approximately 44 million m3 of water per year at the Sant Joan Desp water treatment
facility. According to a water characterization of the Llobregat River by the treatment plant, the mean
concentration of total dissolved solids in the Llobregat is 1,397 mg/L (Table A.1). Another source from
the AGBAR treatment facility reports a similar, although slightly lower mean level of total dissolved
solids at 1,150 mg/L (Pedraz-Yaez 2007). However when one calculates the TDS from the mean
conductivity value (1500 uS/cm) of the Llobregat River with a conversion factor of 0.7 uS/cm per mg/L
TDS, the then TDS value is much higher at 2,140 mg/L.

149.
Ions mg/L
Ammonium (NH4) 0.15
Potassium (K) 31.40
Sodium (Na) 216.90
Magnesium (Mg) 50.80
Calcium (Ca) 162.30
Strontium (Sr) 2.00
Barium (Ba) 0.05
Carbonate (CO3) 1.55
Bicarbonate (HCO3) 370.00
Nitrate (NO3) 11.50
Chloride (Cl) 402.00
Flouride (F) 0.172
Sulfate (SO4) 292.00
Silica (Si02) 8.78
Boron 0.233

Total Dissolved 1,397


Solids
Table A.1 Water quality characteristics of the Llobregat River

ROSA is a computer assisted tool for designing and operating reverse osmosis and nanofiltration systems
(Fig A.1). The software was created to assist water treatment managers operate the desalination systems
provided by the Dow Chemical Company (Dow Chemical 2011).

150.
Figure A.1 Configuration of the AGBAR reverse osmosis system in ROSA.

As described in the Main Text, I ran ROSA with different feed water qualities and then observed the
associated reduction in energy efficiency (kWh/m). I used a conversion factor of 0.7 to transform values
reported in ROSA in Total Dissolved Solids to conductivity. The results of these simulations are
presented in the table below. As presented in the Main Text, we observe an exponential relationship
between TDS values and specific energy efficiency (kWh/m3) with the expression Y=0.315e0.00034x

151.
Savings Savings
per per
reduction reduction
Average in 100 in 100
Net mg/L mg/L TDS
Driving Specific Total TDS [Trendline]
Conductivity Pressure Power Energy Total Energy Energy [Model
TDS (S/cm) (bar) (kW) (kWh/m) (kWh) Expense () Values]

1000 700 6.18 355.12 0.443 19,508,153 1,755,734 39,903 61,074


1100 770 6.05 357.04 0.458 19,951,520 1,795,637 79,806 63,187
1200 840 5.92 358.89 0.474 20,838,255 1,875,443 39,903 65,372
1300 910 5.79 360.69 0.490 21,281,622 1,915,346 79,806 67,633
1400 980 5.65 362.43 0.507 22,168,356 1,995,152 79,806 69,972
1500 1050 5.52 364.13 0.525 23,055,090 2,074,958 79,806 72,392
1600 1120 5.39 365.78 0.543 23,941,824 2,154,764 79,806 74,895
1700 1190 5.26 367.39 0.561 24,828,559 2,234,570 79,806 77,486
1800 1260 5.14 368.96 0.581 25,715,293 2,314,376 79,806 80,166
1900 1330 5.01 370.49 0.601 26,602,027 2,394,182 79,806 82,938
2000 1400 4.88 371.99 0.622 27,488,761 2,473,989 79,806 85,806
Table A.2 TDS, energy, cost relationships.

AGBAR reports that they desalinate approximately 44,336,712 m3 per year and consume nearly 24 GWh
to sustain this desalination. They pay approximately 0.09/kWh for a total expense of approximately 2
million per year only for energy expenses in the reverse osmosis system.

Volume treated by reverse osmosis 44,336,712 m/year


Specific Energy 0.54 kWh/m
Energy Consumption 23,941,824 kWh
Energy Price 0.09 /kWh
Energy Expense 2,154,764 /year
Table A.3 Volumes, energy expenses and costs at the AGBAR treatment facility.

The theoretical minimum amount of energy needed to remove salts from sea water is 1.8 kWh/m3. As of
yet, no technology has been able to reach this theoretical minimum, although new systems are starting to
reach values below 2 kWh/m3 (Elimelech and Phillip 2011).

152.
Appendix B. Electrodialysis Reversal (EDR), Conductivity, & Energy Efficiency
at ATLL

The calculations in the Main Text pertaining to energy use and conductivity at the ATLL facility are
made with observed data. The technological specifications of the EDR system provided by the vendor
(AWWA 1995) also show that lower conductivity in the feed water would reduce energy demand (Table
B.1). Using these values from General Electric, a improvement in feed water quality from <1500 S/cm
to <1000 S/cm would increase energy efficiency from 0.68 kWh/m3 to 0.52 kWh/m3. Maintaining
production volume and energy costs constant, this would imply a reduced expenditure of 425,288 per
year.

Table B.1 Energy consumption (kWh/m3) at ATLL Treatment Facility at different levels of conductivity.
Source
<1000 S/cm <1500 S/cm <2000 S/cm <2500 S/cm
0.52 0.68 0.87 1.02

153.
Appendix C. Heat Fluxes Modeled in the Stream Network Temperature Model

(SNTEMP)
Figure C.1 Heat fluxes in a stream ecosystem. Reproduced from Bartholow (2000)

Figure C.2 Net Heat Flux equation used in SNTEMP Model

154.
Appendix D. Stream Network Model Execution
Model Input Files
The Network Model operates with seven input files and a job control file.
(1) Stream Geometry File KVRFSTR.DAT
(2) Time Period File KVRFTME.DAT
(3) Meteorology File KVRFMET.DAT
(4) Study Node File KVRFSTD.DAT
(5) Hydrology Node File KVRFHDR.DAT
(6) Hydrology Data File KVRFHYD.DAT
(7) Shade File KVRFSHD.DAT
(8) Job Control File KVRFJOB.DAT

Input Data Verification


Before running SNTEMP it is advisable to run a data check program that searches for inconsistencies or
errors in the input tables. The DOS command is: TDATCHK KVRFJOB.DAT KVFCHK.OUT. The
ouput file describes any inconsistencies detected.

Run Command
The model runs with the DOS command SNTEMP. The program will then request the Job Control file:
KVRFJOB.DAT, and the user is asked if any modifications are desired in the Job Control. Modifications
may include the alteration of calibration coefficients or changing the period in which outputs are
requested. If no modifications are made, the Network Model will run through the seven sub-programs:
STRGEM or Stream Geometry, which updates elevations, latitudes and slopes for each node; HYDROL
or Hydrology, which computes missing flows and adds lateral temperatures; METROL or meteorology,
which computes time specific meteorology; REGTWO, which fills in missing temperature values using a
regression model; TRNSPT, which computes water temperatures using physical processes, and VSTATS,
which computes validation statistics (not applicable because no Validation or Calibration nodes used).

Model Output
The model produces eleven output tables. The temperature predictions are in table VIII located in the
output file KVRTRNS.DAT. Users cannot control the names of the output files.

Troubleshooting Issues
SNTEMP would not run unless the input files were formatted precisely. Therefore several issues arose
that prevented the model from running to completion.
(1) Time Period. The model would not run unless each time period (one year) had exactly the same
number of days. This prevented me from using data in 2010 data because I only had data through
November 2010. My period of analysis was revised to 1-January-2004 to 31-December-2009
(later it was reduced further with the need to incorporate the Cardener River for which only data
in 2009 was available). Furthermore, all years needed to have exactly 365 days. The 29th of
February in 2004 and 2008 were eliminated even though data was available.
(2) Regression Program. The regression program that fills in missing temperature values did not run
successfully, and produced a Run Time Error M6201: MATH, which implied that a negative
square root was found in the computations. This was resolved by filling all missing temperature
values with average values from the day prior and after.

155.
Table D.1 Network nodes for the Llobregat Temperature Model

Distance
ID River Node (m) Name Elevation Function UTM_X UTM_Y
1 Llobregat H 57,612 Headwaters Balsareny 285 Headwaters 407107 4633667
2 Llobregat C 49,080 Soler Vicen 175 Change Stream Geometry 409434 4626515
3 Llobregat B 29,810 Llobregat Branch 165 Branch 404459 4615146
4 Cardener H 32,000 Castellgal (Cardener H) 175 Headwaters 403929 4616981
5 Cardener T 29,810 Tributary Cardener 165 Tributary 404459 4615146
6 Llobregat J 29,810 Juncture Cardener 165 Juncture 404459 4615146
7 Llobregat Q 23,932 Castellbell 148 Discharge 405776 4610500
8 Llobregat C 23,010 Bures 141 Change Stream Geometry 404926 4610700
9 Llobregat C 21,752 Borras 141 Change Stream Geometry 404411 4609660
10 Llobregat C 20,462 La Bauma Desviacio 132 Change Stream Geometry 405562 4609180
11 Llobregat C 19,641 La Bauma 127 Change Stream Geometry 405443 4608670
12 Llobregat P 19,547 Castellbell i Vilar WWTP 127 Point Discharge 405366 4608730
13 Llobregat C 18,919 C58 125 Change Stream Geometry 404801 4608890
14 Llobregat C 17,805 Puig i Font 126 Change Stream Geometry 404388 4607880
15 Llobregat C 16,879 Monistrol 119 Change Stream Geometry 403990 4607380
16 Llobregat P 15,054 Monistrol WWTP 109 Point Discharge 404488 4605950
17 Llobregat C 11,048 Cairat 80 Change Stream Geometry 405714 4602760
18 Llobregat C 9,844 La Puda 75 Change Stream Geometry 406672 4602550
19 Llobregat C 6,827 Can Sedo 70 Change Stream Geometry 406079 4599920
20 Llobregat C 2,974 Final Turn 63 Change Stream Geometry 409259 4598810
21 Llobregat D 2,556 D Aigues de Terrasa 62 Diversion 409283 4598400
22 Llobregat E 0 Abrera Treatment Plant 57 End 409757 4596230

156.
Appendix E. Input Files for the Stream Network Model
STUDYFILE [KEY]
River Name Node Distance Upstream Description

STUDYFILE: BALSARENY TO ABRERA 27 JULY 2011


LLOB H 57.612 HEADWATERS BALSARENY
LLOB B 29.81 BRANCH LLOBREGAT
CARD H 32.00 HEADWATERS CARDENER
CARD T 29.81 TRIBUTARY CARDENER
LLOB J 29.81 JUNCTURE CARDENER
LLOB E 0 END ABRERA

HYDROLOGY NODE FILE [KEY]


River name; Node; Output flag; Regression model instructions; Distance Upstream; Description
Hydrology linkage instructions (first line only)

HYDROLOGY NODE FILE: BALSARENY TO ABRERA 27 JULY


LLOB H 57.612 HEADWATERS BALSARENY
LLOB B 29.81 BRANCH LLOBREGAT
CARD H 32.00 HEADWATERS CARDENER
CARD T 29.81 TRIBUTARY CARDENER
LLOB J 29.81 JUNCTURE CARDENER
LLOB Q 23.932 CASTELLBELL
LLOB P 19.547 POINT DISCHARGE CAS WWTP
LLOB P 15.054 POINT DISCHARGE MON WWTP
LLOB D 2.556 AIGUES DE TERRASSA
LLOB E 0 END ABRERA

157.
STREAM SHADE FILE [KEY]
River Name; Node; Output flag; Hydrology model linkage; Local Shade Linkage; Distance Upstream; Description
Site Latitude (Radians); Stream Reach Azimuth (radians); Stream Width (m)
Eastside topographic altitude (radians); vegetation crown (m); height(m); offset (m); vegetation density
Westside topographic altitude (radians); vegetation crown (m); height(m); offset (m); vegetation density
STREAM SHADE FILE: CASTELLBELL TO ABRERA 29 JULY 2011

LLOB H1S2 57.612 C0 HEADWATERS BALSARENY


0.7303 0.1745 25
0.1745 20 20 5 0.15
0.1745 20 20 5 0.15

LLOB C 49.080 CA SOLERVICENC


0.7293 0.0873 30
0.1745 20 20 5 0.15
0.1745 20 20 5 0.15

CARD H1S2 32.000 HEADWATERS CARDENER


0.7292 1.5708 20
0.1745 20 20 5 0.10
0.1745 20 20 5 0.10

LLOB J 29.81 JUNCTURE CARDENER


0.7274 0.1745
0.1745 20 30 5 0.10
0.1745 20 20 5 0.10

LLOB C 23.010 C1 BURES


0.7267 0.0873 30
0.2618 20 20 5 0.2
0.1745 20 20 5 0.05

LLOB C 21.752 C2 BORRAS CANYON


0.7266 -0.7854 30
0.6109 20 20 5 0.1
0.5236 20 20 5 0.8

LLOB C 20.462 C3 LA BAUMA DES


0.7265 -0.2618 25
0.2618 20 20 5 0.25
0.2618 20 20 5 0.1

LLOB C 19.641 C4 LA BAUMA


0.7265 0.7854 30
0.1745 20 20 5 0.4
0.2618 20 20 5 0.4

LLOB C 18.919 C5 C58


0.7263 0.1745 60
0.2967 20 20 5 0.3
0.5236 20 20 5 0.1

LLOB C 17.805 C6 PUIG I FONT


0.7263 0.0873 35
0.2618 20 20 5 0.5
0.5236 20 20 5 0.6

LLOB C 16.879 C7 MONTSERRAT


0.7262 0.7854 25
0.5236 20 20 5 0.2
0.5236 20 20 5 0.2

LLOB C 11.048 C8 CAIRAT


0.7255 -0.349 30
0.4363 20 20 5 0.15
158.
0.4363 20 20 5 0.1

LLOB C 9.844 C9 LA PUDA


0.7255 0.1745 40
0.2618 20 20 5 0.4
0.3491 20 20 5 0.2

LLOB C 6.827 C10 CAN SEDO


0.7252 -0.0873 30
0.2618 20 20 5 0.25
0.3491 20 20 5 0.3

LLOB C 2.974 C11 FINAL TURN


0.7249 0.1745 30
0.1745 20 20 5 0.4
0.1745 20 20 5 0.35

159.
STREAM GEOMETRY FILE [KEY]
River Name; Node; Distance upstream; Description
Site latitude (radians); elevation; Mannings N; Stream Width Coefficient; Width Exponent; Min Shading; Max
Shading; Ground Temperature; Streambed thermal gradient (J/m2/sec/C) default 1.65

STREAM GEOMETRY FILE: BALSARENY TO ABRERA 27 JULY 2011

LLOB H 57.612 HEADWATERS BALSARENY


0.73036 285 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 49.080 CA SOLERVICENC


0.7293 175 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB B 29.81 BRANCH LLOBREGAT


0.72746 165

CARD H 32.00 HEADWATERS CARDENER


0.72749 175 0.035 24.73 0.2061 0.01 0.90 10 1.65

CARD T 29.81 TRIBUTARY CARDENER


0.72746 165

LLOB J 29.81 JUNCTURE CARDENER


0.72746 165 0.035 23.552 0.2061 0.01 0.90 10 1.65

LLOB C 23.010 C1 BURES


0.72678 141 0.035 23.552 0.2061 0.01 0.90 10 1.65

LLOB C 21.752 C2 BORRAS CANYON


0.72661 140 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 20.462 C3 LA BAUMA DES


0.72654 132 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 19.641 C4 LA BAUMA


0.72646 127 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 18.919 C5 C58


0.72649 125 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 17.805 C6 PUIG I FONT


0.72634 126 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 16.879 C7 MONTSERRAT


0.72626 119 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 11.048 C8 CAIRAT


0.72553 80 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 9.844 C9 LA PUDA


0.72550 75 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 6.827 C10 CAN SEDO


0.72510 70 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB C 2.974 C11 FINAL TURN


0.72492 63 0.035 24.73 0.2061 0.01 0.90 10 1.65

LLOB E 0 END ABRERA


0.72452 57 0.035 24.73 0.2061 10 1.65

160.
TIME PERIOD FILE [KEY]
Time Period Name; First Day of Simulation Period; Last Day of Simulation Period; Number of points in time period
average (usually 1); Dust Coefficient for simulation period; Ground reflectivity for simulation period; Calibration
Factors by Time Period: Air temperature calibration constant, Air Temperature calibration coefficient, Wind speed
calibration constant, Wind speed calibration coefficient.

TIME PERIOD FILE


1-Jan 1 1 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
2-Jan 2 2 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
3-Jan 3 3 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
4-Jan 4 4 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
5-Jan 5 5 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
6-Jan 6 6 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
7-Jan 7 7 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
8-Jan 8 8 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000
9-Jan 9 9 1 0.1607 0.2269 0.0000 0.0000 0.0000 0.0000

METEOROLOGY FILE [KEY]


Latitude (radians); elevation (m); Average annual air temp (C); Description
Year; Time period name; Mean air temp; Mean wind speed; Relative humidity; %sunshine (no data); Solar
radiation)

METEOROLOGY FILE: LLOBREGAT WATERSHED - 26 MAY 2011


0.73032 210 14.1 Pont de Vilomara
2009 1-Jan 6.7 0.7 0.916 50.2
2-Jan 6.7 0.5 0.940 31.1
3-Jan 7.7 0.9 0.859 28.8
4-Jan 3.4 1.0 0.865 84.6
5-Jan 2.4 1.7 0.819 72.2
6-Jan 2.1 0.8 0.899 28.3
7-Jan 0.5 1.0 0.863 70.3
8-Jan -1.2 0.8 0.830 84.0
9-Jan 3.0 1.0 0.917 41.2

HYDROLOGY DATA FILE [KEY]


River Name; Node; Distance upstream; Description
Year; Time Period Name; Discharge for time period; Temperature

HYDROLOGY DATA FILE: LLOBREGAT BASIN

LLOB H 57.612 HEADWATERS BALSARENY


2009 1-Jan 4.79 7.25
2-Jan 4.84 7.10
3-Jan 4.71 7.32
4-Jan 4.63 6.66
5-Jan 4.58 5.78

161.
JOB CONTROL FILE [KEY]
Line 1. Title

Line 2. Subtitle

Line 3. Verification Tables. T if requested. F if otherwise.


Tables I-IX; I-III; I; III after M merged; III after O merged; III; IV-V, IV with missing hydrology flows added; V
with lateral flows added; VI; VII; VIII Water Temperature Data; IX Validation statistics; X Job Control; XI;
All skeleton nodes; hydrology nodes; geometry nodes; composite network; B nodes;
C;D;E;H;J;K;M;O;P;Q;R;S;T;V; Hydrology warning message suppression flag; Shade data file present; Global
Shade model used; Regression need flag.

Line 4. General Numeric Information


Number of years of historical data; years of synthetic data; First year (use 0); First year of historical data (use 0);
Number of time periods per year; number of shade nodes; stream geometry nodes; hydrology nodes; study nodes;
total number of nodes; number of nodes requiring regression analysis

Line 5. Node count information


Number of B nodes; C nodes; D nodes; E nodes; H nodes; J nodes; K nodes; M nodes; O nodes; P nodes

Line 6. Node count information and Time Period Output Sequence Numbers
Number of Q nodes; R nodes; S nodes; T nodes; V nodes; Starting year sequence number of output; Last year
sequence number of output; First time period number (always 1); Last time period sequence number (365)

Line 7. Use-Supplied Parameters


User supplied evaporation factor EFA (default =40); User supplied evaporation factor EFB (default =15); User
supplied evaporation factor EFC; User supplied Bowen ratio; 1 st maximum daily air temperature regression
coefficient; 2nd maximum daily air temperature regression coefficient; 3rd maximum daily air temperature
regression coefficient; 4th maximum daily air temperature regression coefficient; Starting time period sequence
number for air temperature correction; Last time period sequence number for air temperature correction

Line 8. Global Calibration Factors


Air temp calibration constant; air temp calibration coefficient; wind speed calibration constant; wind speed
calibration coefficient; humidity calibration constant; humidity calibration coefficient; sunshine calibration
constant; sunshine calibration coefficient; solar calibration constant; solar calibration coefficient

Line 9. Air Temperature Correction Factors


1st elevation; 1st factor; 2nd elevation; 2nd factor; 3rd elevation; 3rd factor; 4th elevation; 4th factor; 5th
elevation; 5th factor

Line 10. Input File Names


Time period file name; Meteorology file name; Skeleton; Stream geometry file name; Study file name

Line 11. Input File Names


Hydrology node file name; Hydrology data file name; Shade file name

Line 12. Spatial Output Request by Stream Name


Local stream name for output; Starting distance for output; Ending distance for output

Lines 13-21. Template (unused)

162.
JOB CONTROL FILE: LLOBREGAT 28 July
TEMPERATURE MODEL
TFFFFFFFFFFFTFFFFFFF FFFFTFFFFFFFFFFFFFFFFTTTT
1. 0. 0. 365. 15. 18. 10. 6. 21. 0.
1. 12. 1. 1. 2. 1. 0. 0. 0. 2.
1. 0. 0. 1. 0. 1. 1. 1. 365. 0.
0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000
0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000
0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000 0.00 0.0000
KVRFTME.DAT KVRFMET.DAT dummy KVRFSTR.DAT KVRFSTD.DAT
KVRFHDR.DAT KVRFHYD.DAT KVRFSHD.DAT
LLOB 57.6 29.8
CARD 29.8 132.0
STR. NAME # 3 0.0 0.0
STR. NAME # 4 0.0 0.0
STR. NAME # 5 0.0 0.0
STR. NAME # 6 0.0 0.0
STR. NAME # 7 0.0 0.0
STR. NAME # 8 0.0 0.0
STR. NAME # 9 0.0 0.0
STR. NAME # 10 0.0 0.0

Output File for Data Verification Program TDATCHK


DATCHK Version 3.4 Results Jul 28 2011 11:28
Job Control File: kvrfjob.dat

Checking Job Control File **************************


Successfully read 21 lines of 21 with 0 warnings.

Checking Time Period File **************************


Successfully read 366 lines of 366 with 0 warnings.

Checking Meteorology File **************************


Successfully read 367 lines of 367 with 0 warnings.

Checking Stream Geometry File **********************


Successfully read 55 lines of 55 with 0 warnings.

Checking Study File ********************************


Successfully read 7 lines of 7 with 0 warnings.

Checking Hydrology Node File ***********************


Successfully read 11 lines of 11 with 0 warnings.

Checking Hydrology Data File ***********************


Successfully read 3671 lines of 3671 with 0 warnings.

Checking Shade File ********************************


Successfully read 76 lines of 76 with 0 warnings.

Cross Checking Nodes *******************************

CARD H 32.0 HEADWATERS CARDENER OK


CARD T 29.8 TRIBUTARY CARDENER OK
LLOB B 29.8 BRANCH LLOBREGAT OK
LLOB C 49.1 CA SOLERVICENC OK
LLOB C 23.0 C1 BURES OK
LLOB C 21.8 C2 BORRAS CANYON OK
LLOB C 20.5 C3 LA BAUMA DES OK
LLOB C 19.6 C4 LA BAUMA OK
LLOB C 18.9 C5 C58 OK
LLOB C 17.8 C6 PUIG I FONT OK
LLOB C 16.9 C7 MONTSERRAT OK
LLOB C 11.0 C8 CAIRAT OK
LLOB C 9.8 C9 LA PUDA OK
163.
LLOB C 6.8 C10 CAN SEDO OK
LLOB C 3.0 C11 FINAL TURN OK
LLOB D 2.6 AIGUES DE TERRASSA OK
LLOB E 0.0 END ABRERA OK
LLOB H 57.6 HEADWATERS BALSARENY OK
LLOB J 29.8 JUNCTURE CARDENER OK
LLOB P 19.5 POINT DISCHARGE CAS WWTP OK
LLOB P 15.1 POINT DISCHARGE MON WWTP OK
LLOB Q 23.9 CASTELLBELL OK

Node cross check completed with 0 warnings.

Cross Checking Time Periods ************************

Time period check completed with 0 warnings.

164.
Appendix F. Field Work for Shading Estimates
The field work familiarized me with important stream features. For example, I learned that the stream
temperature gage at Castellbell was not at the location of the UTM coordinates provided. Neither was the
temperature gage, which was several hundred meters downstream from the Castellbell discharge gage. 124
Below is the template form used during the collection of field data.
Node/Waypoint

Name

UTM X

UTMY

Radians

Comments

C Node: Topographic Vegetation


Number & Description

UTM X

UTMY

Latitude (radians)

Stream width (m)

Stream reach azimuth (radians)

West Side East Side


Topographic Altitude (degrees) Topographic Altitude (degrees)

Vegetation crown diameter (m) Vegetation crown diameter (m)

Vegetation height: water to top (m) Vegetation height: water to top (m)

Offset: trunk to water (m) Offset: trunk to water (m)

Continuity of vegetation (%) Continuity of vegetation (%)

Screening factor (%) Screening factor (%)

Min/Max Density (%) Min/Max Density (%)

124
For photographs during field work along the Llobregat River see:
https://picasaweb.google.com/jhoney/LlobregatRiverSegment?authkey=Gv1sRgCIyulbGwtI613wE#
165.
Appendix G. Calibration of SNTEMP Model
The uncalibrated model predicted stream temperatures that were too warm throughout the year. This
indicated that a cool effect in the watershed was not accounted for. This might be from the Gavarresa
tributary that flows into the Llobregat. The Gavarresa is the only significant tributary not accounted for in
the model because temperature data was not available.

The air temperature coefficient was the only parameter used to calibrate the model. Calibration was
coded into the time period file (KVRFTME.DAT) to allow for seasonal adjustments. During the winter,
from November to February, the air temperature calibration coefficient was 0.25. During the peak
summer months of June to September it was 0.9. In the remaining months the air temperature calibration
coefficient was 0.5.

Figure G.1 Observed and predicted temperature values with calibrated and uncalibrated Stream Network
Temperature Model for the Llobregat River in 2009.

166.
Appendix H. Full Results for SNTEMP
In the period in which EDR modules operated (March to October) the calibrated model predicted stream
temperatures with a mean error of 0.53C, while for the entire year the model error was 1.23C.

Figure H.1 Network Temperature Model Error Distribution.

Restorable Restorable Restorable All Areas All Areas No Coverage


Areas to Areas to Areas to to 80% to 100%
40% 60% 80%
Mean -0.264 -0.513 -0.764 -1.068 -1.427 0.424
Mean (C/km) -0.005 -0.009 -0.013 -0.019 -0.025 0.007
Minimum -0.580 -1.090 -1.610 -2.190 -2.920 -0.160
Maximum 0.070 0.150 0.220 0.340 0.460 0.850
SD 0.147 0.286 0.427 0.612 0.821 0.247
Table H.1 Summary statistics of temperature changes from scenarios that modified riparian shading.

167.
5% 25% 50% 100% 200%
Mean -0.048 -0.227 -0.293 -0.585 -0.773
Mean C/km -0.001 -0.004 -0.005 -0.010 -0.014
Minimum -0.130 -0.580 -1.100 -2.400 -3.210
Maximum 0.100 0.510 0.870 1.030 1.220
SD 0.034 0.152 0.268 0.577 0.736
Table H.2 Summary statistics of temperature changes from scenarios that increased stream discharge.

Restorable40% Restorable60% Restorable80% 80% 100% No Coverage


Annual -0.26 -0.51 -0.76 -1.07 -1.43 0.42
March-October -0.34 -0.65 -0.97 -1.38 -1.84 0.54

Table H.3 Temperature change (C) associated with vegetation scenarios.

5% 25% 50% 100%


Annual -0.04 -0.18 -0.35 -0.70
March-October -0.04 -0.19 -0.36 -0.65
Table H.4 Temperature change (C) associated with discharge scenarios

Threshold Crossings
I correct for bias in my tabulation of threshold crossings with the regression equation that gave
me the relationship between modeled and observed observations (Fig 4.4). This quantified the
bias of my model.

Equation (1): Modeled temperature=0.8068(observed temperature) + 4.335.

I solved for observed temperature:

(2) Observed temperature = (Modeled Temperature 4.355)/ 0.8068.

These updated temperature (or predicted temperature) account for the model bias and therefore
were more appropriate for my estimation of threshold crossings.

168.
Temperature Threshold (C)
12 14 16 18 20 22 24 26 Total Euros/ year
Days

Restorable 40% 4 12 6 1 5 15 9 5 57 57,000

Restorable 60% 8 20 14 4 8 22 17 10 104 104,500

Restorable 80% 12 26 21 7 15 35 26 14 156 156,000

100% 30 39 28 17 40 64 44 20 283 283,000

No Coverage -2 -7 -9 -8 -5 -12 -23 -13 -79 (79,000)

Table H.5 Avoided threshold crossings (in days) and their estimated economic value (Euros) associated
with each of the five shading scenarios increases in stream shading.

Temperature Threshold (C)


12 14 16 18 20 22 24 26 Total Euros/
Days Year
5% 2 1 2 0 0 0 1 1 7 6,250

25% 7 9 5 0 2 9 4 4 40 40,000

50% 8 17 8 2 5 15 10 10 75 75,000

100% 15 26 11 4 12 21 13 18 120 85,000

Table H.6 Avoided threshold crossings (in days) and their estimated economic value (Euros) from
increases in stream discharge.

169.
Llobregat River Temperature Profile
27 July 2009
25
24
23

Stream Temperature (C)


22
21 No Coverage
20 Baseline
19 Restorable to 40%
18 100% Discharge
17
100% Coverage
16
15
50 40 30 20 10 0
Distance upstream (Km)

Figure H.2 Temperature Profile 27 July 2009.

Llobregat River Temperature Profile


10 July 2009
25
24
Stream Temperature (C)

23
22 No Coverage
21
Baseline
20
Restorable to 40%
19
18 25% Discharge
17 100% Discharge
16 100% Coverage
15
60 50 40 30 20 10 0
Distance upstream (Km)

Figure H.3 Temperature Profile 10 July 2009.

170.
Figure H.4 Stream temperature according to the revegetation scenarios.

171.
Appendix I. Notes on Scale Selection for SNTEMP
The results are sensitive to the length of stream segment studied, since most of the thermal heating occurs
in the first half of the stream segment between Balsareny and the stream juncture with the Cardener. As
noted in the Main Text, temperatures rise more rapidly at first, and then more gradually in the second half
of the segment. In this second half, between Castellbell and Abrera temperatures rise an average only
1.35C over this 29 km reach. This is an average of 0.06C/km, which is the lowest of the reaches
studied. In contrast, the upper segment between Sallent and Castellbell, temperatures rise on average
0.11C/km -- nearly double the thermal change in the lower half.

Three factors probably amplify the effect of stream shading in the upper reaches of the watershed. First
the stream width is narrower, which allows for shading to cover a larger proportion of the stream. Second,
the stream volume is smaller, and larger streams are less sensitive to changes in shading conditions
(Seedang et al. 2008, Chan et al. 1998). Third, temperatures higher in the watershed are cooler to begin
with, so shading helps the stream remain cool. Once the warming has taken place, shading has a much
smaller impact on downstream values. This implies that restoration in the upper half of the stream
segment would yield greater thermal protection than restoration in the lower reaches.

It is worth mentioning that the temperature rise on the Llobregat River is relatively small when compared
to other streams globally. The rate of increase for small streams has been reported as 0.6C/km
(Zwieniecki and Newton 1999), intermediate streams 0.2C/km (Cassie 2006) while larger streams 0.09
C/km (Torgerson et al. 2001). This would place the Llobregat with an average thermal gain per
kilometer closer to large streams, even though its discharge is more like an intermediate, or small stream.

Figure I.1 Observed stream temperature values in the Llobregat River during 2009.
172.
Figure I.2 Temperature differences between upstream and downstream gages.

Figure I.3 Mean Temperature Rise (C/km) between downstream and upstream gages.

173.
Figure I.4 Mean Temperature Rise (C/km) between downstream and upstream gages.

174.
Appendix J. Stream Segment Model (SSTEMP)

Prior to using the SNTEMP model as presented in the Main Text, I used the Stream Segment Model
(SSTEMP) to make a preliminary approximation regarding the potential impact of increased shading and
discharge on stream temperature. The Segment Model was designed to be a simplified and scaled down
version of SNTEMP (Bartholow 2004). The Segment Model includes a Windows interface to that allows
users to enter input variables and estimate downstream temperatures for a one day period. The Segment
Model uses the same mathematical equations as the original Network Model but does not allow for
mixing (tributaries, diversions or point source contributions), nor does it permit changes to geographic,
topographic or vegetative conditions within the segment.

The stream segment studied in the SSTEMP model was shorter (23.932 km) than the studied reach in the
Main Text. The segment began at the discharge measurement gage in Castellbell i el Vilar and ended at
the ATLL drinking water treatment facility. This segment begins 53 km downstream of the Baells dam
and 4.5 km downstream from the Llobregats confluence with the Cardener River. This reach is fed by
only small tributaries, most of which flow only during storm events. This segment has nine water
diversions to hydroelectric generators. These diversions are, in order moving downstream: Burs, Borrs,
La Bauma, Puig i Font, Comes, Gomis, El Cairat, Can Sed, and Catex Moli. Two waste water treatment
facilities discharge along this segment, one at Castellbell i el Vilar, and a second at Monistrol de
Montserrat. The stream segment ends 29 km upstream of its terminus point in the Mediterranean.

Just like SNTEMP, SSTEMP calculates the heat gained or lost from a parcel of water as it passes through
a stream segment. The model simulates heat flux processes such as convection, conduction, evaporation,
direct solar radiation (short wave), heat from the air (long wave radiation), and radiation back from the
water (see Main Text).

I collected data with a mean daily time step for the period 1 January 2004 to 15 November 2010
(n=2,526). Meteorological data was obtained from the same weather station in the town of el Pont de
Vilomara (406310, 4617994 UTM), operated by the Catalan Meteorological Service, and 7.5 kilometers
north of the river segment. Hydrologic and stream temperature data was downloaded from the Catalan
Water Agency (ACA).

The Segment Model was designed to produce mean daily outputs one day at a time (Fig M.1). A utility
allowed me to run several observations in a single operation, but only if all input parameters had data.
When removing days in which information was incomplete, the final data set included 1,470 days.

175.
Figure J.1 The user interface for the Segment Model (SSTEMP). Users adjust conditions concerning
time of year, hydrology, geometry, meteorology, and shade. The model predicts mean temperature values
at the end of the stream segment given these conditions (lower right).

176.
Figure J.2 Stream reach studied for the Stream Segment Temperature Model (SSTEMP).

SSTEMP Results
The Segment Model predicted downstream temperature with a mean error of -1.53 C (Fig M.3). Over
40% of the predicted temperatures were within 1 C of the observed values.

177.
Figure J.3 The error distribution of the Segment Model during 2004-2010. Error measured in C.

The Segment Model underestimated temperature values at the ATLL-Abrera water treatment facility,
especially during summer months. The most accurate results were during 2009 and 2010, while the error
increased in the earlier years (Fig M.4).

SSTEMP Temperature Estimates in


Llobregat at Abrera
30.00

25.00
Degrees Celsius

20.00

15.00
Abrera
10.00 Modeled

5.00

0.00

Figure J.4 Observed temperature (blue) and modeled values (red) with the Stream Segment Temperature
model. Gaps show periods where input parameters were not available in the ACA water temperature data
set, most notably in late 2007 and early 2008.

178.
The Segment Model makes bold assumptions concerning the entire stream segment, and yet it succeeded
in estimating downstream temperatures under a wide range of meteorological and hydrological
conditions. The observed and modeled scatter plot shows a tight correlation (Pearson Correlation
Coefficient= 0.975, R2= 0.9249) albeit with a consistent underestimate (Fig M.5).

Figure J.5 The Segment Model underestimated downstream temperature values across a wide range of
seasonal conditions.

The consistent underestimation by Segment Model suggests that certain factors contributing to stream
heating which were not being incorporated into the model. These sources of heat may include the influx
of urban waste water, or the diversion of water into the hydroelectric channels.

The Segment Model also includes a tool to conduct a sensitivity analysis. The Segment Model was most
sensitive to inflow temperature, air temperature, relative humidity and solar radiation (Fig M.6). The
models sensitivity to these parameters is consistent with other stream temperature models (Bartholow
1989). In contrast, the parameters of most interest for this research, total shading and segment inflow,
showed that they had little influence over temperature. A variation in 10% of both of these parameters
made no difference in output temperature, according to the Segment Model sensitivity analysis.

179.
Figure J.6 Sensitivity Analysis for Segment Model showing that output results are most sensitive to
inflow temperature, air temperature, relative humidity and solar radiation.

Riparian Shading in Segment Model


The Segment Model had a baseline shade estimate of 10% for the entire segment. To test the possible
influence of riparian vegetation on stream temperature, I ran the Segment Model with the same conditions
but with an increased total shading estimate of 30%. This modification simulated temperature values with
an increase in riparian shading.

However the simulated temperatures influenced my modeled results, not observed values. To estimate the
real temperature reduction on the observed values, I used the linear regression equation derived from
Figure M.5 that plotted the relationship between observed and modeled values.

Equation (1) Y=0.9249x 0.3154

I ran the modeled temperature values through equation (1) under both normal conditions and simulated
conditions. The difference between these temperature values was my new estimate of temperature
reductions. I then subtracted the temperature reductions from the observed temperature values. Finally, I
counted the number of days the treatment facility would have remained below the temperature thresholds
of 12C, 14C, 16C, 18C, 20C, 22C, 24C and 26 C.

The simulation suggests that increased shading moderated stream temperature sufficiently to prevent a
temperature threshold crossing on 280 of 1,470 days (19%), with an average of 65.5 days per year. The
simulation also suggests that an increase in shading would prevent the more threshold crossings in
warmer conditions, especially around 24 C.

180.
Figure J.7 Number of days water treatment facility will avoid crossing the temperature thresholds with an
increase in riparian shading to 30%.

Simulating Increased Instream Flows in the Segment Model


Discharge volumes in the Llobregat River are regulated by three dams: Baells, Sant Pon, and Llosa del
Cavall. In addition, water flow is diverted to generate hydroelectric power at abandoned textile mills.
These diversions reduce flows in the main channel of the Llobregat. The low flows are disruptive to the
aquatic ecosystem and may contribute to thermal heating. Currently the Catalan Water Agency is taking
steps to regulate minimum instream flows and oblige the owners of the hydroelectric generators to reduce
their diversions.

With the Segment Model I simulated temperature values with increased flow through the main channel.
For a first approximation, I simulated the doubling of discharge values. On a day by day basis, the model
could also generate stream temperature profiles under various discharge conditions.

181.
Figure J.8 Number of days in which simulated stream temperatures would not pass critical threshold
values due to an increase in stream discharge.

The simulated increase in discharge moderated thermal heating, preventing threshold crossings on 1,012
days (68%), averaging 253 crossings avoided per year. Using the same assumptions as before, this
simulated increase in discharge could generate savings of approximately 253,000 per year for the ATLL-
Abrera water treatment facility.

Economic Value of Ecosystem Services


Each of the nine EDR modules at the ATLL-Abrera water treatment facility costs approximately 1,000
to operate per day.125 We can approximate the value of the shading service (V Shading or Discharge) by
estimating the number of days (N) in which we avoid the cost an additional module (CModule):

VShading or Discharge = NDays*CModule

The simulation suggests the ATLL-Abrera facility would avoid crossing the temperature threshold on
average 70 days per year. An EDR module costs approximately 1,000 per day to operate, therefore the
shading service under these conditions would be 70,000 per year.

125
The modules consume approximately 12,000 kWh per day, at a rate of 0.09 /kWh.
182.
Appendix K. Stream Network Model from Castellbell to Abrera

Before running the final version of the Stream Network Model presented in the Main Text, I ran a Stream
Network Model that considered a smaller reach between Castellbell and Abrera (23.9 km). This model
also originally included the nine water diversions by the various hydropower generators. This iteration of
the Network Model consisted of 34 nodes when including all the hydroelectric diversions (Appendix 10
Figure 1). When the hydroelectric diversions were removed, the simplified network contained 13 nodes.
Data on stream discharge must be provided for nodes Headwaters (H), Diversion (D), Point Source (P),
and End (E). Temperature data must be provided for Headwaters (H), and Point Source (P).

Figure K.1 A schematic representation of the complete and simplified Stream Network Model between
the headwaters in Castellbell i el Vilar to the ATLL Water Treatment Plant in Abrera.

183.
Figure K.2 Restorable stream segments. Areas in Red have less than 30% stream cover and have the
topographic conditions that would make riparian restoration plausible.

Results SNTEMP Model: Castellbell to Abrera


The simplified Network Model overestimates downstream temperature by an average of 2.14 C. During
the months in which the EDR system operates, from March to October, the overestimate is reduced to
1.37 C. The overestimate is particularly strong in the winter months, which are less relevant for this
study.

184.
Figure K.3 Modeled and observed temperature values during the period 2004-2009

Figure K.4 A comparison of observed temperature values with modeled values predicted by Network
Model at the end of the stream network. The Network Model shows a tight correlation but also
overestimates values.

185.
Temperature Profile of Stream Segment on
Llobregat River
20
18
Temperature (C)

16
14
12
10
8
6
4
2
0
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34
Figure K.5 Temperature profile of the SNTEMP model moving downstream from the first node H to the
end node. The spike at point 3 is attributable to a node located after a water diversion. At this location
there is little discharge in the main channel and the model estimates a high stream temperature. This high
temperature disappears once the diverted water is returned to the main channel. I have verified that the
rise in temperature on node 3 only occurs when the diversion upstream takes nearly all the flow from the
main channel.

Table K.1 Changes in Riparian Shading


Scenario Vegetation Change Mean Mean
Temperature Temperature
Change (C) Change (C)
From March to
October

1 Restorable segments increased to 40% -0.08 -0.10


coverage (most realistic scenario)
2 All segments at least 40% coverage -0.14 -0.17

3 All segments increased to 80% vegetation -0.55 -0.69


coverage
4 All segments increased to 100% vegetation -0.77 -0.96
coverage
5 Stream vegetation removed or 0% coverage 0.32 0.40

186.
Figure K.6 Stream Temperature Change from Modification in Riparian Vegetation for the SNTEMP
model between Castellbell and Abrera.

187.
Table K.2 Network Nodes for the SNTEMP model between Castellbell and Abrera

ID River Node Name Elevation Function UTM_X UTM_Y NodeID Upstream (m)
1 Llobregat H Headwaters 148 Headwaters 405776 4610500 H 23932.10
2 Llobregat D Bures 147 Diversion 405385 4610500 D1 23552.10
3 Llobregat C Bures 141 Change Stream 404926 4610700 C1 23010.50
4 Llobregat R R Bures 141 Geometry
Return 404673 4610510 R2 22662.80
5 Llobregat C Borras 141 Change Stream 404411 4609660 C2 21752.60
6 Llobregat D Can Borras 139 Geometry
Diversion 404486 4609650 D2 21674.30
7 Llobregat R R Borras 132 Return 405116 4609280 R1 20942.40
8 Llobregat C La Bauma Desviacio 132 Change Stream 405562 4609180 C3 20462.90
9 Llobregat D La Bauma 132 Geometry
Diversion 405547 4609130 D3 20427.90
10 Llobregat R R La Bauma 127 Return 405661 4608820 R3 19907.40
11 Llobregat C La Bauma 127 Change Stream 405443 4608670 C4 19641.30
12 Llobregat P Castellbell i Vilar 127 Geometry
Point Discharge 405366 4608730 P1 19547.90
13 Llobregat C WWTP
C58 125 Change Stream 404801 4608890 C5 18919.50
14 Llobregat C Puig i Font 126 Geometry
Change Stream 404388 4607880 C6 17805.60
15 Llobregat D Puig i Font 126 Geometry
Diversion 404375 4607890 D4 17789.10
16 Llobregat R R Puig i Font 119 Return 403957 4607530 R4 17036.90
17 Llobregat C Monistrol 119 Change Stream 403990 4607380 C7 16879.10
18 Llobregat D Comes 118 Geometry
Diversion 404210 4607020 D5 16466.10
19 Llobregat R R Comes 112 Return 404317 4606920 R5 16303.30
20 Llobregat D Colonia Gomis 111 Diversion 404401 4605970 D6 15152.80
21 Llobregat P Monistrol WWTP 109 Point Discharge 404488 4605950 P2 15054.30
22 Llobregat R R Gomis 106 Return 404403 4605240 R6 14142.60
23 Llobregat D El Cairat 106 Diversion 405380 4603030 D7 11480.40
24 Llobregat R R El Cairat 80 Return 405464 4602940 R7 11378.10
25 Llobregat C Cairat 80 Change Stream 405714 4602760 C8 11048.90
26 Llobregat C La Puda 75 Geometry
Change Stream 406672 4602550 C9 9844.90
27 Llobregat D Can Sedo 75 Geometry
Diversion 406198 4600790 D8 7786.30
28 Llobregat R R Can Sedo 70 Return 406068 4600150 R8 7050.40
29 Llobregat C Can Sedo 70 Change Stream 406079 4599920 C10 6827.40
30 Llobregat D Catex Moli 70 Geometry
Diversion 407124 4599570 D9 5716.10
31 Llobregat R R Catex Moli 63 Return 409111 4598890 R9 3138.20
32 Llobregat C Final Turn 63 Change Stream 409259 4598810 C11 2974.30
33 Llobregat D D Aigues de Terrasa 62 Geometry
Diversion 409283 4598400 D10 2556.90
34 Llobregat E ATLL-Abrera 57 End 409757 4596230 E 0.00

188.
Appendix L. Riparian Vegetation and Restoration Costs
Starting in 2006, the Catalan Water Agency has collaborated with municipalities to restore riparian
habitat along the Cardener and Llobregat Rivers. Projects have been executed at 16 sites in which
invasive species of cane (arundo donax) has been removed, and native trees and shrubs were planted.
The planted tree species include willows (Salix alba), black alder (Alnus glutinosa), white poplar
(Populus alba), narrow leafed ash (Fraxinus angustifolia) and black poplar (Populus nigra). They also
have planted native shrubs such as the rosemary willow (Salix elaeganos), the purple willow (Salix
purpurea), the common dogwood (Cornus sanguinea), the European Privet (Ligustrum vulgaris), the
common Hawthorn (Crataegus monogyna), and the black thorn (Prunus spinosa).

As an example of what is planted at these riparian restoration projects, the table below provides a list of
what was planted at Monistrol de Montserrat.

Common Name Species Trees Planted Height


White Poplar Populus alba 30 16-27 m.
Narrow leafed Ash Fraxinus angustifolia 15 20-30 m.
Willow Salix alba 15 10-30 m.
Table L.1 Tree Species planted along the Llobregat River in Monistrol de Montserrat.

The average heights for these tree species reach up to 27-30 meters. However my shading estimates were
conservative and only assumed heights of 20 m.

Figure L.1 Signage of the riparian restoration project at Monistrol de Monsterrat.

189.
A cost-benefit analysis make assumptions pertaining to: (a) discount rate, (b) tree growth rate, (c)
shading-value relationship.

(a) Discount rate. Cost benefit analysis often use discount rates in the range of 4-6%. In this case, I have
chosen a discount rate of 4%. Higher discount rates imply that present generations place less value on
future costs and benefits. Lower discount rates imply that present generations place a greater value future
costs and benefits.

(b) Tree growth rate. I found tree growth rates for riparian species in another study that estimated the
effect of riparian restoration on stream temperature (Wanatabe et al. 2005). In the Llobregat River, the
restoration projects also planted species of cottonwoods, making these growth rates particularly useful.
These growth curves tell us that cottonwoods will reach a height of 20 m at some point between 20 and 25
years. Since the trees planted in these restoration projects are already several years old, I assume that 20
years after planting, the trees will be 20 m in height (Fig L.2). For my shading model, 20 m is full
maturity, because that was the maximum height used in the model.

Figure L.2 Tree growth of cottonwoods and other vegetation used for riparian restoration (Wanatabe et
al. 2005).

The temperature model assumed a maximum height of only 20 m. The tree growth rate used in the
cost/benefit analysis was adjusted accordingly, so that after 20 years, the trees reached 100% of its
expected height.

(c) Shading-value relationship


One critical assumption concerns how the shading effects are translated into economic value. In this case,
we assume that the economic value of the shading effect is exactly proportional to the percent maturity of
the tree (% Growth with 20m being 100% maturity). To calculate the precise relationship between
economic value and ecosystem structure I would need to run the model for each year with the trees at
different heights. For each year, I would need to calculate the number of avoided threshold crossings.
Note that I am assuming a linear relationship between ecosystem growth and net benefit for the water
treatment plant. As pointed out in the introduction, this linear assumption may not always be the case
(Aburto-Oropeza et al. 2008, Barbier et al. 2008, Koch et al. 2009).

190.
Cost per km 60,150
Discount Rate 0.04
Scenario 1 Scenario 2 Scenario 3 Scenario 4
Restorable Areas to 40% Restorable Areas to 60% Restorable Areas to 80% 100 % Coverage
Reach Length (km) 18 27 36 57.6
Initial Investment 1,082,700 1,624,050 2,165,400 3,464,640
% Tree Growth Year Future Value Present Value Future Value Present Value Future Value Present Value Future Value Present Value
0.05 1 2,850 2,740 5,200 5,000 7,800 7,500 14,150 13,476
0.1 2 5,700 5,270 10,400 9,615 15,600 14,423 28,300 25,669
0.15 3 8,550 7,601 15,600 13,868 23,400 20,803 42,450 36,670
0.2 4 11,400 9,745 20,800 17,780 31,200 26,670 56,600 46,565
0.27 5 15,390 12,649 28,080 23,080 42,120 34,620 76,410 59,869
0.3 6 17,100 13,514 31,200 24,658 46,800 36,987 84,900 63,354
0.35 7 19,950 15,160 36,400 27,661 54,600 41,492 99,050 70,393
0.4 8 22,800 16,660 41,600 30,397 62,400 45,595 113,200 76,618
0.45 9 25,650 18,021 46,800 32,881 70,200 49,322 127,350 82,091
0.56 10 31,920 21,564 58,240 39,345 87,360 59,017 158,480 97,293
0.58 11 33,060 21,475 60,320 39,183 90,480 58,774 164,140 95,969
0.6 12 34,200 21,361 62,400 38,975 93,600 58,462 169,800 94,551
0.65 13 37,050 22,251 67,600 40,599 101,400 60,898 183,950 97,553
0.7 14 39,900 23,041 72,800 42,040 109,200 63,060 198,100 100,054
0.77 15 43,890 24,371 80,080 44,466 120,120 66,698 217,910 104,818
0.8 16 45,600 24,346 83,200 44,421 124,800 66,632 226,400 103,716
0.85 17 48,450 24,873 88,400 45,382 132,600 68,073 240,550 104,951
0.9 18 51,300 25,323 93,600 46,204 140,400 69,305 254,700 105,833
0.95 19 54,150 25,702 98,800 46,895 148,200 70,342 268,850 106,393
1 20 57,000 26,014 104,000 47,464 156,000 71,196 283,000 106,660
Undiscounted Benefits 605,910 1,105,520 1,658,280 3,008,290
Net Benefits 361,683 659,913 989,869 1,592,497
Net Cost 1,082,700 1,624,050 2,165,400 3,464,640
Net Present Value - 721,017 - 964,137 - 1,175,531 - 1,872,143
Percent Recovery 33% 41% 46% 46%
Table L.2 Cost-Benefit Analysis for each of the restoration scenarios.

191.
REFERENCES
Aburto-Oropeza, O., E. Ezcurra, G. Danemann, V. Valdez, J. Murray and E. Sala. 2008. Mangroves in the Gulf of
California increase fishery yields. Proceedings of the National Academy of Sciences 30: 10456-10459.
ACA. 2010. Programa de mesures del Pla de gesti del districte de conca fluvial de Catalunya. Annex II. Llistat
dactuacions. Generalitat de Catalunya. Departament de Medi Ambient i Habitatge. Barcelona, Spain.
ACA. 2009. Projecte constructiu de les actuacions destinades a la reducci de limpacte ambiental del runam inactiu
de Vilafruns. Generalitat de Catalunya. Departament de Medi Ambient i Habitage. Barcelona, Spain.
ACA. 2006. Agncia Catalana de l'Aigua. Memria 2006. Generalitat de Catalunya. Departament de Medi Ambient
i Habitatge. Barcelona, Spain.
ACA. 2006b. Agncia Catalana de l'Aigua. Pla Zonal dimplementaci de cabals demanteniment a la conca de lAlt
Ter. Anlisis de centrals hidroelctriques. Novembre 2006. Barcelona, Spain.
ACA. 2003. Projecte de Restauraci Integral de la Vall Salina, Cardona. Memria. Agncia Catalana de lAigua.
Generalitat de Catalunya. Departament de Medi Ambient i Habitatge. Barcelona, Spain.
Acua, V. and K. Tockner. 2009. Surface-subsurface water exchange rates along alluvial river reaches control the
thermal patterns in an Alpine river network. Freshwater Biology 54: 306-320. Doi:10/1111/j.1365-
2427.2008.02109.x
Acua, V., A. Wolf, U. Uehlinger and K. Tockner. 2008. Temperature dependence of stream benthic respiration in
an Alpine river network with relevance to global warming. Freshwater Biology 53: 2076-2088.
Adams, P., D.E. Nelson, S. Yamada, W. Chmara, R.G. Jensen, H. J. Bohnert and H. Griffiths. 1998. Tansley Review
No 97 Growth and development of Mesembryanthemum crystallinum (Aizoaceae) New Phytologist 138:
171-190.
Alabern, et al. 1991. Historia de la Ciutat de Manresa (1900-1950). Caixa de Manresa. Manresa, Spain.
Aldom, I. 2007. La lluita per l'aigua a Catalunya. De l'us, abs a la gesti integral (1900-2007). Pags Editors.
Lleida, Spain.
Aldy, J. E., J. Hrubovcak and U. Vasavada. 1998. The role of technology in sustaining agriculture and the
environment. Ecological Economics 26: 81-96.
American Water Works Association (AWWA). 1995. Electrodialysis and Electrodialysis Reversal. AWWA Manual
M38. Denver, CO.
Arbol, C. 2011. LACA obt 400 milions dels bancs pels provedors. ARA. 17 November 2011. 24.
http://www.ara.cat/premium/cronica/LACA-obte-milions-bancs-proveidors_0_592740782.html, accessed
on 22 April 2012.
Armsworth, P.R., A. N. Armsworth, N. Compton, P. Cottle, I. Davies, B.A. Emmett, V. Frandrich, M. Foote, K.J.
Gaston, P. Gardiner, T. Hess, J. Hopkins, N. Horsley, N. Leaver, T. Maynard and D. Shannon. 2010. The
ecological research needs of business. Journal of Applied Ecology 47: 235-243.
Armsworth, P.R., K. M. A. Chan, G. C. Daily, P. R. Ehrlich, C. Kremen, T. H. Ricketts, and M. A. Sanjayan. 2007.
Ecosystem-service science and the way forward for conservation. Conservation Biology 21(12): 1383-
1384.
Arnau i Reitg, R. 1981. La mineria del Bages: Una visi retrospectiva. In XXVI Assamblea Intercanvis comarcals
destudiosos a Manresa. Centre dEstudis del Bages. Ajuntament de Manresa. Manresa, Spain. 53-58.
Arrojo Agudo, P. 2003. El Plan Hidrolgico Nacional: Una cita frustrada con la historia. Integral: Barcelona,
Spain.
Arrojo Agudo, P., and J.M. Naredo. 1997. La gestin del agua en Espaa y California. Coleccin Nueva Cultura del
Agua. Bakeaz: Bilbao.
Arrow, K., G. Daily, P. Dasgupta, S. Levin, K. G. Maler, E. Maskin, D. Starrett, T. Sterner and T. Tietenberg. 2000.
Managing Ecosystem Resources. Environmental Science and Technology. 34: 1401-1406.
ATLL. 2008. Estudi per a loptimitzaci econmica-sanitria del funcionament conjunt del tractament convencional
I la instal.laci delectrodilisis reversible a lETAP del Llobregat (T.M. Abrera). Planificaci del servei de
producci daigua per al consum hum. Aiges Ter Llobregat. 17 september 2008.
Avlonitis, S.A. 2002. Operational water costs and productivity improvements for small-size RO desalination plants.
Desalination. 142: 295-304.
Ayala, E., P. Serra, P. Villa and M. Vilajosana. 1983. Una mina, un poble. Centre Excursionista de Catalunya.
Editorial Montblanc-Martin: Barcelona, Spain.
192.
Badia, E. 1996. La sal, suport duns pobles. Angle Editorial: Manresa, Spain.
Bai, Y., C. Zhuang, Z. Ouyan, H. Zheng and B. Jiang. 2011. Spatial characteristics between biodiversity and
ecosystem services in a human-dominated watershed. Ecological Complexity 8: 177-183.
Bakker, K. 2002. From state to market? water mercantilizacin in Spain. Environment and Planning A 34: 767-790.
Balcells, A. 1998. De la crisi del rgim autonmic a laixecament military del 1936. In Histria de Catalunya.
Volum XI. 1586-1600. Salvat Editores: Barcelona, Spain.
Barbier, E.B., E.W. Koch, B.R. Silliman, S.D. Hacker, E. Wolanski, J. Primavera, E.F. Granek, et al. 2008. Coastal
ecosystem-based management with nonlinear ecological functions and values. Science 319: 321-323.
Barbier, E.B. 2007. Valuing ecosystem services. Economic Policy. January: 177-229.
Bartholow, J.M. 2004. SSTEMP for Windows: The Stream Segment Temperature Model (Version 2). [Available
online at http://www.fort.usgs.gov/Products/Publications/10016/10016.pdf]
Bartholow, J.M. 2003. Modeling uncertainty: Quicksand for water temperature modeling. Hydrological Science and
Technology 19(1-4): 221-232.
Bartholow, J.M. 2000. The Stream Segment and Stream Network Temperature Models: A Self-Study Guide. US.
Geological Survey. US Department of the Interior. Version 2.0 March 2000. Open File Report 99-112. US
Geological Survey computer model and documentation. [Available online at
http://www.fort.usgs.gov/products/software/SNTEMP/]
Bartholow, J.M. 2000b. Estimating cumulative effects of clearcutting on stream temperatures. Rivers 7(4): 284-297.
Bartholow, J.M. 1991. A Modeling Assessment of the Thermal Regime for an Urban Sport Fishery. Environmental
Management. 15(6): 833-845.
Bartholow, J.M. 1989. Stream Temperature Investigations: field and analytic methods. U.S. Fish and Wildlife
Service. Instream Paper No. 13. Biological Report 89: 17. Washington D.C.
Bates, W.T., C. Bartels and K. Lai. 2010. Brackish water RO and NF operations on high TDS feed waters.
American Membrane Technology Association. [online:
http://www.membranes.com/docs/papers/New%20Folder/BRACKISH%20WATER%20RO%20AND%20
NF%20OPERATIONS%20AMTA%202010%20Final%20071610.pdf]
BenDor, T.K. and M.W. Doyle. 2010. Planning for Ecosystem Service Markets. Journal of the American Planning
Association 76 (1): 59-72.
BenDor, T.K., and N. Brozovic. 2007. Determinants of Spatial and Temporal Patterns in Compensatory Wetland
Mitigation. Environmental Management 40: 349-364.
Bennett, E.M., G.D., Peterson and L.J. Gordon. 2009. Understanding relationships among multiple ecosystem
services. Ecology Letters 12: 1394-1404.
Beschta, R.L. 1997. Riparian Shade and Stream Temperature: An Alternative Perspective. Rangelands 19(2): 25-28.
Bishop, J. and N. Landell-Mills. 2002. Forest environmental services: an overview. In Pagiola, S., J. Bishop and N.
Landell-Mills, editors. 2002. Selling forest environmental services market-based mechanisms for
conservation and development. Earthscan Publications: London, UK. 15-24.
Boada, M. 1995. Rafael Puig i Valls (1845-1920) Precursor de l'educaci ambiental i dels espais naturals protegits.
Departament de Medi Ambient, Generalitat de Catalunya. Barcelona, Spain.
Boardman, A.E. 1996. Cost benefit analysis: Concepts and practice. Prentice Hall: Upper Saddle River, NJ.
Bohnert, H.J. and J.C. Cushman 2000. The ice plant cometh: Lessons in abiotic stress tolerance. Journal of Plant
Regulation 19: 334-346.
Bolaos, A. 2004. Aiges de Barcelona. Histria dels reptes per al subministrament daigua. In Prat, N. and E. Tello.
El Baix Llobregat: histria i actualitat ambiental dun riu. Centre dEstudis Comarcals del Baix Llobregat.
214-225.
Bolund, P. and S. Hunhammar. 1999. Ecosystem services in urban areas. Ecological Economics 29(2): 293-301.
Boyd, J. and S. Banzhaf. 2007. What are ecosystem services? The need for standardized environmental accounting
units. Ecological Economics 63:616-626.
Boyd, M. and D. Sturdevant. 1997. The scientific basis for Oregons Stream Temperature Standard. Common
Questions and Straight Answers. Oregon Department of Environmental Quality, Salem, Oregon.
Brauman, K.A., G.C. Daily, T. Kaeo Duarte and H.A. Mooney. 2007. The nature and value of ecosystem services:
An overview highlighting hydrologic services. Annual Review of Environment and Resources 32: 67-98.
Brechin, G.A. 1999. Imperial San Francisco: Urban power, earthly ruin. California studies in critical human
geography 3. University of California Press: Berkeley.

193.
Brenan, G. 1943. The Spanish Labyrinth (12th Edition ed.). Cambridge University Press: Cambridge.
Brouwer, R. and M. Hofkes. 2008. Integrated hydro-economic modeling: Approaches, key issues and future
research directions. Ecological Economics 66(1): 16-22.
Brozovic, N., J. Honey-Ross and D. W. Schneider (in prep). Technological Change and The Value of Ecosystem
Services.
Brozovic, N. and W. Schlenker. 2011. Resilience, uncertainty, and the role of economics in ecosystem management
Ecological Economics 70(4): 627-640.
Butturini, A., T. J. Battin and F. Sabater. 2000. Nitrogen in stream sediment biofilms: the role of ammonium
concentrations and DOC quality. Water Resources 34(2): 629-639.
Cabeza Diaz, R. 1997. L'aigua, un recurs universal i escas. Beta Editorial: Barcelona.
Cabin, R.J. 2007. Science driven restoration: a square grid on a round earth? Restoration Ecology 15:1-7.
Carol, M. 1991. El Llobregat: Un Cam daigua. Langwerg Editions. Diputaci de Barcelona. Barcelona, Spain.
Carpenter, S. R., H. A. Mooney, J. Agard, D. Capistrano, R.S. DeFries, S. Daz, T. Dietz, et al. 2009. Science for
managing ecosystem services: Beyond the millennium ecosystem assessment. Proceedings of the National
Academy of Sciences 106(3): 1305-12.
Carpenter, S.R., R. DeFries, T. Dietz, H.A. Mooney, S. Polasky, W.V. Reid and R.J. Scholes. 2006. Millennium
Ecosystem Assessment: Research needs. Science 314: 257-8.
Cassie, D. 2006. Thermal regime of rivers: a review. Freshwater Biology 51: 1389-1406.
Castree, N. 2005. Nature. Routledge. London: U.K.
Chan, K.M.A., L. Hoshizaki and B. Klinkenberg. 2011. Ecosystem services in conservation planning: Targeted
benefits vs. Co-Benefits or Costs? PLoS ONE 6(9):e24378. doi:10.1371/journal.pone.0024378.
Chan, K.M.A., R. M. Pringle, J. Ranganathan, C.L. Boggs, Y.L. Chan, P.R. Ehrlich, P.K. Haff, N.E. Heller, K. Al-
Khafaji and D.P. Macmynowski. 2007. When Agendas Collide: Human Welfare and Biological
Conservation. Conservation Biology 21(1): 59-68.
Chan, K.M.A., M.R. Shaw, D.R. Cameron, E.C. Underwood, and G.C. Daily. 2006. Conservation planning for
ecosystem services. PLoS Biology 4 (11): 2138-52.
Chen, N., H. Li and L. Wang. 2009. A GIS-based approach for mapping direct use value of ecosystem services at a
country scale: Management implications. Ecological Economics 68(11): 2768-2776.
Chen, Y.D., S.C. McCutcheon, D.J. Norton and W.L. Nutter. 1998. Stream Temperature Simulation of Forested
Riparian Areas II. Model Applications. Journal of Environmental Engineering 124(4): 316-328.
Chen, Z.M., G. Q. Chen, B. Chen, J. B. Zhou, Z. F. Yang and Y. Zhou. 2009. Net ecosystem services value of
wetland: Environmental economic account. Communications in Nonlinear Science and Numerical
Simulation 14(6) (6): 2837-43.
Chichilnisky, G., and G. Heal. 1998. Economic returns from the biosphere. Nature 391: 629-630.
Child, M.F. 2007. The Thoreau Ideal as a Unifying Thread in the Conservation Movement. Conservation Biology
23(2): 241-243.
Cioc, M. 2002. The Rhine. An Eco-biography, 1815-200. University of Washington Press: Seattle, WA.
Clark, W.C. 2007. Sustainability science: a room of its own. Proceedings of the National Academy of Sciences 104:
1737-1738.
Clever, M., F. Jordt, R. Knauf, N. Rabiger, M. Rudebusch and R. Hilder-Scheibel. 2000. Process water production
from river water by ultrafiltration and reverse osmosis. Desalination 131:325-336.
Clevland, C.J. and M. Ruth 1997. When, where and by how much do biophysical limits constrain the economic
process? A survey of Nicholas Georgescu-Roegens contribution to ecological economics. Ecological
Economics 22: 203-223.
Codina, J. 1971. El delta del llobregat i barcelona. gneres i formes de vida dels segles XVI al XX. Hores de
catalunya. Edicions Ariel: Esplugues de Llobregat.
Codina, J. 1971b. Inundacions al delta del llobregat. Episodis de la historia 147-148. Rafael Dalmau: Barcelona.
Consell Comarcal del Bages and Phragmites SL. 2011. Informe sobre leradicaci de la canya Arundo donax.
Report to ACA (internal document).
Consell Comarcal del Bages and Phragmites SL. 2010. Informe sobre leradicaci de la canya Arundo donax.
Report to ACA (internal document).
Constanz, J. 1998. Interactions between stream temperature, streamflow, and groundwater exchanges in alpine
streams. Water Resources Research 34(7): 1609-1615.

194.
Corbera, E., C. Gonzlez Soberanis and K. Brown. 2009. Institutional dimensions of Payment for Ecosystem
Services: An analysis of Mexicos carbon forestry programme. Ecological Economics 68: 743-761.
Doi:10.1016/j.ecolecon.2008.06008.
Corbera, E., N. Kosoy and M. Martnez-Tuna. 2007. The equity implications of marketing ecosystem services in
protected areas and rural communities: case Studies from Meso-America. Global Environmental Change
17: 365-380.
Corlyvan, M., J. Justus and H.H. Regan. 2010. The natural environment is valuable but not infinite. Conservation
Letters doi.10.1111/j.1755-263x.2010.0018x
Corporacin Metropolitana de Barcelona. 1986. Usos agrcolas de los mrgenes y delta del Llobregat. Barcelona.
Cortner, H.J., M.G. Wallace, S. Burke and M.A. Moote. 1998. Institutions matter: the need to address the
institutional challenges of ecosystem management. Landscape and urban planning 40: 159-166.
Costanza, R. 2006. Nature: ecosystems without commodifying them. Nature 443: 749.
Costanza, R. 2003. A vision of the future of science: reintegrating the study of humans and the rest of nature.
Futures 35: 651-671.
Costanza, R., B.S. Low, E. Ostrom and J.A. Wilson 2001. Ecosystems and human systems: A framework for
exploring the linkages. In: Costanza, R. B.S. Low, E. Ostrom and J.A. Wilson. Institutions, Ecosystems and
Sustainability. Lewis Publishers: Washington D.C.
Costanza, R., et al. 1997. The value of the worlds ecosystem services and natural capital. Nature 387: 253-260
Costanza, R. and H.E. Daly. 1992. Natural Capital and Sustainable Development. Conservation Biology 6(1): 37-46.
Countant, CC. 1985. Stripped Bass, Temperature, and Dissolved-Oxygen A speculative hypothesis for
environmental risk. Transactions of the American Fisheries Society 114(1): 31-61.
Cowling, R.M., B. Egoh, A.T. Knight, P.J. OFarrel, B. Reyers, M. Rouget, D.J. Roux, A. Welz and A. Wilhem-
Rechman. 2008. An operational model for mainstreaming ecosystem services for implementation.
Proceedings of the National Academy of Sciences (PNAS) 105(28): 9483-9488.
Crittenden J., R.R. Trussell, D.W. Hand, K.J. Howe and G. Tchobanoglous. 2005. Water treatment: principles and
design. John Wiley & Sons: Hoboken, New Jersey.
Cronon, W. 1995. Uncommon ground: toward reinventing nature. W.W. Norton & Co.: New York.
Crossman, N.D. and B.A. Bryan. 2009. Identifying cost-effective hotspots for restoring natural capital and
enhancing landscape multifunctionality. Ecological Economics 68(3): 654-68.
Daily, G.C., S.Polasky, J. Goldstein, P.M. Kareiva, H.A. Mooney, L. Pejchar, T.H. Ricketts, J. Salzman, and R.T.
Shallenberger. 2009. Ecosystem services in decision making: Time to deliver. Frontiers in Ecology and the
Environment 1: 21-8.
Daily, G.C. and P.A. Matson. 2008. Ecosystem Services: From theory to implementation. Proceedings of the
National Academy of Sciences 105: 9455-9456.
Daily, G.C. and Ellison, K. 2002. The New Economy of Nature. Island Press. Washington D.C.
Daily, G.C., T. Sderqvist, S. Aniyar, K. Arrow, P. Dasgupta, P. R. Ehrlich, C. Folke, et al. 2000. The value of
nature and the nature of value. Science 289: 395-396.
Daily, G.C. 1997. Natures Services. Island Press: Washington D.C.
Daily, G.C. 1997. Introduction: What are ecosystem services? In: G. C. Daily. Nature's services: Societal
dependence on natural ecosystems 1-10. Island Press: Washington, D.C.
Dearmont, D., B. McCarl, and D. Tolman. 1998. Costs of water treatment due to diminished water quality: A case
study in Texas. Water Resources Research 34(4): 849-53.
Delgado, J., P. Llorens, G. Nord, I.R. Calder and F. Gallart. 2010. Modelling the hydrologic response of a
Mediterranean medium-sized headwater basin subject to land cover change: the Cardener River (NE
Spain). Journal of Hydrology 383: 125-134.
Denzin, N. 1978. Sociological Methods: A Sourcebook. McGraw Hill: NY. 2nd edition.
De Sarra, M. 1935. El abastecimiento de aguas de Barcelona y los yacimientos de sles potsicas. Qumica e
Industria XII (134): 51-56. (Biblioteca de Catalunya)
Daz, S., S. Lavorel, F. de Bello, F. Qutier, K. Grigulis and T. M. Robson. 2007. Incorporating plant functional
diversity effects in ecosystem service assessments. Proceedings of the National Academy of Sciences 104:
20684-20689.
Diputaci de Barcelona, no date. Projecte Riu Verd a Monistrol de Montserrat. Public Signage on Llobregat River
in Monistrol de Montserrat. Photograph taken May 2010.

195.
Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework
for Community action in the field of water policy. [Available online http://eur-
lex.europa.eu/LexUriServ/LexUriServ.do?uri=CELEX:32000L0060:EN:NOT] Accessed 23 April 2012.
Dobbs, T.L. and J. Pretty. 2008. Case study of agri-environmental payments: The United Kingdom. Ecological
Economics 65(4): 765-775.
Dow Chemical. 2011. Reverse Osmosis System Analysis (ROSA) 7.2.7. Available online
[http://www.dowwaterandprocess.com/support_training/design_tools/rosa.htm]
Duran Ventosa, L. 1914. Memoria sobre alguns problemes ques presenten en el riu Llobregat. Diputaci Provincial
de Barcelona. Barcelona (Biblioteca de Catalunya).
Ehrlich, P. R. and H.A. Mooney. 1983. Extinction, Substitution, and Ecosystem Services. Bioscience 33: 248-254.
Eigenbrod, F., V.A. Bell, H.N. Davies, A. Heinemeyer, P.R. Armsworth and K. J. Gaston. 2011. The impact of
projected increases in urbanization on ecosystem services. Proceedings of the Royal Society B 278: 3201-
3208.
Eigenbrod, F., B.J. Anderson, P.R. Armsworth, A. Heinemeyer, S.F. Jackson, M. Parnell, C.D. Thomas and K.J.
Gaston. 2009. Ecosystem service benefits of contrasting conservation strategies in a human-dominated
region. Proceedings of the Royal Society B 276: 2903-2911.
Eisenhardt, K.M., and M.E. Graebneer 2007. Theory building from cases: opportunities and challenges. Academy
of Management Journal 50(1):25-32.
Elimelech, M. and W. A. Phillip. 2011. The future of seawater desalination: energy, technology and the
environment. Science 333: 712-717.
Escaler, I. 2007. Medium and long term water resources modeling as a tool for planning and global change
adaptation. Application to the Llobregat Basin. Life+ Environment Policy and Governance. European
Union Funding Proposal.
Estevan, A. and N. Prat. 2006. Alternativas para la gestin del agua en Catalua: Una visin desde la perspectiva de
la nueva cultura del agua. Fundacin Nueva Cultura del Agua. Zaragoza and Bakeaz: Bilbao, Spain.
European Union (EU). 2000. Directive 2000/60/EC of the European Parliament and of the Council of 23 October
2000 establishing a framework for Community action in the field of water policy. Official Journal of the
European Communities.
Expansin 2011. Mas-Colell abre la puerta a la privatizar empresas pblicas. Section: Catalua. 19 January. Pg 7.
Fbrega Enfedaque, A. 2009. Cum Grano Salis: La Sal i la Potassa a Sria 1185-1982. Ajuntament de Sria &
Iberpotash.
Farber, S., R. Costanza, D.L. Childers, J. Erickson, K. Gross, M. Grove, C.S. Hopkinson, J. Kahn, S. Pincetl, A.
Troy, P. Warren and M. Wilson. 2006. Linking Ecology and Economics for Ecosystem Management.
Bioscience 56(2): 121-133.
Fernandez-Turiel, J.L., D. Gimeno, J.J. Rodriguez, M. Carnicero and F. Valero. 2003. Spatial and Seasonal
Variations of water quality in a Mediterranean catchment: The Llobregat River (NE Spain). Environmental
Geochemistry and Health 25: 453-474.
Ferraro, P.J., and A. Kiss. 2002. Direct payments to conserve biodiversity. Science 298: 1718-1719.
Ferraro, P.J., and R. D. Simpson. 2002. The cost-effectiveness of conservation payments. Land Economics 78: 339-
353.
Ferret, J. 2006. Els orgens del processos de salinitzaci de les aiges subterrnies de la conca del Llobregat per les
explotacions de sals potssiques (1923-1936). Unpublished manuscript.
Ferret, J. 1985. Laprofitament de les aiges subterrnies del Delta del Llobregat. Comunitat dUsuaris DAiges de
lrea Oriental del Delta del Riu Llobregat. LHospitalet, Spain.
Fisher, B. et al. 2011. Measuring, modeling and mapping ecosystem services in the Eastern Arc Mountains of
Tanzania. Progress in Physical Geography 35(5): 595-611.
Fisher, B., R. K. Turner and P. Morling. 2009. Defining and classifying ecosystem services for decision making.
Ecological Economics S 68: 643-653.
Fisher, B. et al. 2008. Ecosystem Services and Economic Theory: Integration for Policy Relevant Research.
Ecological Applications. 18(8): 2050-2067.
Forman, R.T.T. 2004. Mosaico territorial para la regin metropolitana de Barcelona (Land Mosaic for the Greater
Barcelona Region: Planning a Future). Editorial Gustavo Gili: Barcelona.
Forman, R.T.T. 1995. Land Mosaics. Cambridge University Press: Cambridge, United Kingdom.

196.
Gandy, M. 2002. Concrete and Clay: Reworking Nature in New York City. Urban and industrial environments. MIT
Press: Cambridge, Mass.
Garriga, J. 2011. Lleonard Carcol, Director de lAgncia Catalana de lAigua, LATLL no s ven, per es podria
externalitzar la gesti del seu servei. LEconmic. 12 November 2011: 4.
Garriga, J. 2011b. Aigues Pbliques Turbulentes. LEconmic. 12 November 2011: 2-3.
Goeller, H.E. and A.M. Weinberg. 1976. The Age of Substitutability. Science 20: 683-689.
Goldstein, J.H. 2007. Paying for Conservation in Human Dominated Landscapes. Doctoral Dissertation. Stanford
University.
Gmez-Baggethun, E.R. de Groot, P. L. Lomas and C. Montes. 2010. The history of ecosystem services in
economic theory and practice: From early notions to markets and payment schemes. Ecological Economics
69: 1209-1218.
Gopal, K., S Swarupa Tripathy, J.L. Bersilon and S. Prabha Dubey. 2007. Chlorination by products their
toxicodynamics and removal from drinking water. Journal of Hazardous Materials 140: 1-6.
Gorostiza, S., J. Honey-Ross and R. Lloret (in prep). Rius de Sal: Una visi histrica de la salinitzaci dels rius
Llobregat i Cardener durant el segle XX.
Gorostiza, S. 2010. El conflicto salino en el suministro de agua a Barcelona (1925 1940). Encuentro Cientfico
Salud y ciudades en Espaa, 1880-1940. Condiciones ambientales, niveles de vida e intervenciones
sanitarias. 8 y 9 de julio de 2010, Barcelona.
Graczyk, D.J. and W.C. Sonzogni 1991. Reductions of dissolved-oxygen concentrations in Wisconsin streams
during summer runoff. Journal of Environmental Quality 20(2): 445-451.
Greenlee, L.F., D.F. Lawler, B.D. Freeman, B. Marrot, P. Moulin. 2009. Reverse Osmosis desalination: Water
sources, technology, and todays challenges. Water Research 43: 2317-2348.
Grimm, N.B., R.W. Sheibley, C.L. Crenshaw, C.N. Dahm, W.J. Roach and L.H. Zeglin. 2005. N retention and
transformation in urban streams. Journal of the North American Benthological Society 24(3): 626-642
doi: 10.1899/0887-3593.
Guerrero, R.M.R., del Rio Zuloaga, R., Gonzlez., 1979. Presa de la Baells. Revista de Obras Pblicas. Diciembre
1079-1095.
Guo, Zhongwei, Xiangming Xiao and Dianmo Li. 2000. An assessment of ecosystem services: Water flow
regulation and hydroelectric power production. Ecological Applications 10 (3): 925-36.
Guterl, F. 2005. Investing in Green. Newsweek. June 1: 36.
Hanley, N. and A. Z. Black. 2006. Cost benefit analysis and the Water Directive Framework in Scotland. Integrated
Environmental Assessment and Management 2: 156-165.
Hein, L., K. van Koppen, R. S. de Groot and E C. van Ierland. 2006. Spatial scales, stakeholders and the valuation of
ecosystem services. Ecological Economics 57 (2) (5/1): 209-28.
Hendrick, R. and J. Monahan. 2003. An assessment of water temperatures of the Entiat River, Washington using the
Stream Network Temperature Model (SNTEMP). Report for the Entiat WRIA Planning Unit (EWPU).
[Available online at www.cascadiacd.org/files/documents/SNTEMP_FinalDraft_Sept03.pdf]
Hester, E.T. and M.N. Gooseff. 2010. Moving beyond the banks: hypohreic restoration is fundamental to restoring
ecological services and functions of streams. Environmental Science and Technology 44: 1521-1525.
Holling, C. S. 1978. Adaptive environmental assessment and management. Wiley IIASA international series on
applied systems analysis Chichester. Wiley: New York.
Hope,D., C. Gries, W. Zhu, W.F. Fagan, C.L. Redman, N.B. Grimm, A.L. Nelson, C. Martin and A. Kinzig. 2003.
Socioeconomics drive urban plant diversity. Proceedings of the National Academy of Sciences 100(15):
8788-8792.
Howarth, R. B. and S. Farber. 2002. Accounting for the value of ecosystem services. Ecological Economics 41: 421-
429.
Huesemann M.H. 2001. Can pollution problems be effectively solved by environmental science and technology? An
analysis of critical limitations. Ecological Economics 37: 271-287.
Hughes, R. 1992. Barcelona. Knopf: New York.
Jaffe, A. B., R.G. Newell and R.N. Stavins. 2005. A tale of two market failures: Technology and environmental
policy. Ecological Economics 54: 164-176.
Jansson, A., C. Folke, J. Rockstrom and L. Gordon. 1999. Linking freshwater flows and ecosystem services
appropriated by people: The case of the Baltic sea drainage basin. Ecosystems2 4: 351-66.

197.
Johnson, S.L. 2004. Factors influencing stream temperature in small streams: substrate effects and a shading
experiment. Canadian Journal of Fisheries and Aquatic Science 61: 913-923.
Johnson, S.L. 2003. Stream temperature: scaling of observations and issues for modeling. Hydrological Processes
17: 497-499.
Kauffman, J. B., R.L. Beschta, R.L, Ottiny and D. Lytjen. 1997. An ecological perspective of riparian and stream
restoration in the western United States. Fisheries 22: 12-24.
Karagiannis, I.C. and P.G. Soldatos, 2008. Water desalination cost literature: review and assessment. Desalination
223: 448-456.
Kareiva, P., S. Watts, R. McDonald and T. Boucher. 2007. Domesticated Nature: Shaping Landscapes and
Ecosystems for Human Welfare. Science 316: 1866-69.
Koch, E.W., E. B. Barbier, B. R. Silliman, D. J. Reed, G. M. E. Perillo, S. D. Hacker, E. F. Granek, et al. 2009. Non-
linearity in ecosystem services: Temporal and spatial variability in coastal protection. Frontiers in Ecology
and the Environment 1: 29-37.
Kosoy, N. and E. Corbera. 2010. Payments for Ecosystem Services as Commodity Fetishism. Ecological Economics
69 (6): 12281236
Kremen, C. 2005. Managing ecosystem services: what do we need know about their ecology? Ecology Letters 8:
468-479 doi:10.1111/j.1461-0248.2005.00751x
Kremen, C. and R.S. Ostfeld. 2005. A call to ecologists: Measuring, analyzing, and managing Ecosystem services.
Frontiers in Ecology and the Environment 3: 540-548.
Kreuter, U.P., H.G. Harris, M.D. Matlock and R.E. Lacey. 2001. Change in ecosystem service values in the San
Antonio area, Texas. Ecological Economics 39(3): 333-346. DOI: 10.1016/S0921-8009(01)00250-6
Kroll, F., F. Mller, D. Haase and N. Fohrer. 2012. Rural-urban gradient analysis of ecosystem services supply and
demand dynamics. Land Use Policy: 521-535.
Krutilla, J.V. 1969. Conservation reconsidered. American Economic Review 57: 777-786.
Larson, L.L. and S. L. Larson. 1996. Riparian Shade and Stream Temperature: A Prespective. Rangelands 18(4):
149-152.
Latorre, X. 1995. Histria de laigua a Catalunya. Labecedari: Premia de Mar, Spain.
LeBlanc, R.T., R. D. Brown and J. E. FitzGibbon. 1997. Modeling the effects of land use change on water
temperature in unregulated urban streams. Journal of Environmental Management 49: 445-469.
LEconmic 2011. El preu de no cobrar b laigua. Editorial. LEconmic. 12 November 2011. Pg 12.
Liu, S., R. Costanza, S. Farber and A. Troy. 2010. Valuing ecosystem services: Theory, practice, and the need for a
transdisciplinary synthesis. Annals of the New York Academy of Sciences 1185: 54-78.
Liu, J., S. Li, Z. Ouyang, C. Tam and X. Chen. 2008. Ecological and socioeconomic effects of China's policies for
ecosystem services. Proceedings of the National Academy of Sciences 105(28): 9477-82.
Lloret, R. 2004. La qualitat de laigua del riu Llobregat. Un factor limitant del passat, un element clau per al futur.
In Prat, N and Tello, E. El Baix Llobregat: histria i actualitat ambiental dun riu. Centre destudis
Comarcals del Baix Llobregat: 92-141.
Lpez, G. 1926. Las Aguas de Barcelona : impugnacin a la memoria de Los Servicios de la Sociedad General de
Aguas de Barcelona. (Biblioteca de Catalunya)
Lundy, L. and R. Wade. 2011. Integrating science to sustain urban ecosystems. Progress in Physical Geography
35(5): 653-669.
Luque, F. 2008. Tratamiento del Agua del Ro Llobregat en la ETAP de Sant Joan Desp (Barcelona) por
Membranas de Ultrafiltracin y smosis Inversa. Asociacin Espaola de Desalacin y Reutilizacin
(AEDyR) VII Congreso AEDyR. December 3-5.
Marsh, G. P. 1874. The Earth as Modified by Human Action. A New Edition of Man and Nature. Sampson Low,
Marston Low, and Searle: London. Republished by Elibron Classics 2006.
Martn Pasqual, J.M. 2007. Aigua i Societat a Barcelona entre les dues Exposicions 1888-1929. Doctoral
Dissertation. Departament dHistria Contempornea i Moderna. Universitat Autnoma de Barcelona.
Available online: http://www.tesisenxarxa.net/TDX-1213107-105345/
Martnez-Alier, J. 2002. The Environmentalism of the Poor. Edward Elgar: Northampton, MA, USA.
Masats i Llover, J. 1997. Histria de la Indstria Txtil a Castellbell i el Vilar. Centre d'Estudis del Bages. Manresa,
Spain.

198.
Mastin, M.C. 2008. Effects of potential future warming on runoff in the Yakima River Basin, Washington: U.S.
Geological Survey Scientific Investigations Report 2008: 5124-5136.
McCauley, D.J. 2006. Selling out on nature. Nature 443: 27-8.
McCully, P. 2001. Silenced Rivers. The ecology and politics of large dams. Zed Books: New York.
McDonnell, M.J. and S.T.A. Pickett 1990. Ecosystem structure and function along urban-rural gradients: An
unexploited opportunity for ecology. Ecology 71(4): 1232-1237.
McHarg, I. 1969. Design with Nature. Doubleday/Nat. Hist: Garden City, NJ.
Meehan, W.R. 1970. Some effects of shade cover on stream temperature in southeast Alaska. USDA Forest Service
Res. Note PNW 113. Pacific Northwest Forest and Range Experiment Station.
Mehan, G.T. 2009. Congressional testimony before the Subcommittee on Water Resources and Environment of the
House Committee on Transportation and Infrastructure on Sustainable Water Management. 4 February.
Melosi, M.V., 1999. The Sanitary City: City: Urban Infrastructure in America from Colonial Times to the Present.
The Johns Hopkins University Press, Baltimore, MD.
Meyer, J.L., M. J. Paul and W. K. Taulbee. 2005. Stream Ecosystem Function in Urbanizing Landscapes. Journal of
the North American Benthological Society 24(3): 602-612.
Millennium Ecosystem Assessment. 2005. Ecosystems and human well-being: Current states and trends. Island
Press: Washington D.C.
Millennium Ecosystem Assessment. 2003. Ecosystems and human well-being: A framework for assessment. Island
Press: Washington D.C.
Miralda, A and P. Vall. 2002. Model de desenvolupament territorial per a les colnies del Llobregat. Fundaci
CaixaManresa, Manresa.
Moberg, F. and P. Rnnbck. 2003. Ecosystem services of the tropical seascape: interactions, substitutions and
restoration. Ocean & Coastal Management 46: 27-46.
Montenegro, M. 2008. The market force of nature. SEED Magazine. April 27, 2008.
Mooney H. and P.R. Ehrlich. 1997. Ecosystem Services: A Fragmentary History. In: Daily G. C., ed. Nature's
Services: Societal dependence on natural ecosystems, 11-19. Island Press: Washington, DC.
Mujeriego, R. 2006. Abastament daigua des del Baix Llobregat nord: Diagnosi per a la millora de la qualitat.
Agncia Catalana de lAigua, Aiges Ter-Llobregat, Direcci General de Salut Pblica. Generalitat de
Catalunya. Barcelona, Spain.
Munn, A. and N. Prat. 2004. Defining River Types in a Mediterranean Area: A methodology for the
implementation of the EU Water Framework Directive. Environmental Management 34(5):711-729
Muoz-Pia, C., A. Guevara, J. M. Torres and J. Braa. 2008. Paying for the hydrological services of Mexico's
forests: Analysis, negotiations and results. Ecological Economics 65 (4) (5/1): 725-36.
Muradian, R., E. Corbera, U. Pascual, N. Kosoy and P.H. May. 2010. Reconciling theory and practice: An
alternative conceptual framework for understanding payments for environmental services. Ecological
Economics 69: 1202-1208.
Naidoo, R., A. Balmford, R. Costanza, B. Fisher, R.E. Green, B. Lehner, T.R. Malcom and T.H. Ricketts. 2008.
Global mapping of ecosystem services and conservation priorities. Proceedings of the National Academy of
Sciences 105: 9495-9500.
Naidoo, R. and T.H. Ricketts. 2006. Mapping the economic costs and benefits of conservation. PloS Biology. 4(11):
2153-2164. e360 DOI:10.137/journal.pbio.0040360
National Research Council. 2005. Valuing Ecosystem Services: Toward Better Environmental Decision Making.
National Academy Press: Washington, D.C.
National Research Council. 2000. Watershed management for potable water supply: Assessing the New York City
strategy. National Academy Press Washington, D.C.
National Research Council. 1999. New Strategies for Americas Watersheds. National Academy Press: Washington,
D.C.
Nelson E.G. et al. 2009. Modeling multiple ecosystem services, biodiversity conservation, commodity production,
and tradeoffs at landscape scales. Frontiers in Ecology and the Environment 1: 4-11.
New York Times. 2004. Save the watershed. Editorial Desk. Sunday Edition Section 4; Page 14; Column 1.
Nikolaou, A.D., S.K. Golfinopoulous, G.B. Arhonditsis, V. Kolovoyiannis and T.D. Lekkas. 2004. Modeling the
formation of chlorination by-products in river waters with different quality. Chemosphere 55: 409-420.

199.
Norgaard, R.B. 2010. Ecosystem services: From eye-opening metaphor to complexity blinder. Ecological
Economics 69 (6): 1219-1227.
Norton, G.E. and A. Bradford. 2009. Comparison of two stream temperature models and evaluation of potential
management alternatives for the Speed River, Southern Ontario. Journal of Environmental Management
90: 866-878.
Oberndorfer, E., J. Lundholm, B. Bass, R.R. Coffman, H. Doshi, N. Dunnett, S. Gaffin, M. Kohler, K.K.Y. Liu, and
B. Rowe. 2007 Green roofs as urban ecosystems: Ecological structures, functions, and services. Bioscience.
57(10): 823-833 doi: 10.1641/B571005.
Oliver, B., ,J.J. Alonso and J.R. Catalan. Estudio Hidrolgico del Ro Llobregat. 1971. CEIA, Litocolor, SA:
Barcelona.
ONeill, B. 2004. Cassandra/Cornucopian Debate. International Encyclopedia of the Social & Behavioral Sciences:
1525-1529. doi:10.1016/B0-08-043076-7/04145-0.
Otero, N. and A. Soler. 2002. Sulphur isotopes as tracers of the influence of potash mining in groundwater
salinisation in the Llobregat Basin (NE Spain). Water Research 36: 3989-4000.
Pagiola, S. 2008. Payments for environmental services in Costa Rica. Ecological Economic 65: 712-24.
Pagiola, S., J. Bishop and N. Landell-Mills, editors. 2002. Selling forest environmental services market-based
mechanisms for conservation and development. Earthscan Publications: London, UK.
Palmer, M.A. and S. Filoso. 2009. Restoration of Ecosystem Services for Environmental Markets. Science 325: 575-
576.
Pedraz Yaez, G. 2007. Proyecto y obras de la mejora del tratamiento de agua por osmosis inversa en la ETAP de
Sant Joan Desp. Document 1. Memoria. Aguas de Barcelona: Barcelona, Spain.
Peterson, G.D., S.R. Carpenter and W.A. Brock. 2003. Uncertainty and the Management of Multistate Ecosystems:
An Apparently Rational Route To Collapse Ecology 84: 14031411.
Pickett, S.T.A., M.L. Cadensasso, J.M. Grove, C.H. Nilon, R.V. Pouyat, W.C. Zipperer and R. Costanza. 2001.
Urban Ecological Systems: Linking Terrestrial, Ecological, and Socioeconomic Components of
Metropolitan Areas. Annu. Rev. Ecol. Syst. 32: 127-57.
Pigou, A.C. 1920. The Economics of Welfare. Macmillan and Co.: London, United Kingdom. 4th Edition.
Power, M.E., N. Brozovic, C. Bode and D. Zilberman. 2005. Spatially explicit tools for understanding and
sustaining inland water ecosystems. Frontiers in Ecology and the Environment (1): 47-55.
Prats, J., R. Val, J. Armengol and J. Dolz. 2010. Temporal variability in the thermal regime of the lower Ebro River
(Spain) and alteration due to anthropogenic factors. Journal of Hydrology 387: 105-118.
Pritchard, L., C. Folke and L. Gunderson. 2000. Valuation of Ecosystem Services in Institutional Contexts.
Ecosystems 3: 36-40.
Puig i Valls, R. 1904. El Llobregat: sus cuencas alta, media y baja y obras indispensables que hay que realizar en
ellas, para conseguir que las inundaciones sean cada vez menos temibles, y las aguas normales ms
constantes, con aumentos de riqueza pblica y particular. Memorias de la Real Academia de Ciencias y
Artes de Barcelona, tecera poca IV (40): 524-536. Barcelona.
Puig i Valls, R. 1890. El Llobregat: aguas y montes. Revista de Montes. Ao XIV. Nm 325-327. Madrid. Real
Academia de Ciencias y Artes de Barcelona. pgs 357-366, 377-388, 427-439.
Raudsepp-Hearne, C., G.D. Peterson and E.M. Bennett, 2010. Ecosystem bundles for analyzing tradeoffs in diverse
landscapes. Proceedings of the National Academy of Sciences 107: 11140-11144.
Raudsepp-Hearne, C., G.D. Peterson, M. Teng, E.M. Bennett, T. Holland, K. Benassaiah, G.K. MacDonald and L.
Pfeifer. 2010b. Untangling the Environmentalists Paradox: Why is human well-being increasing as
ecosystem services degrade? BioScience 60: 576-589.
Real Decreto 140/2003. Establishment of Spanish Drinking Water Quality Standards. 7 February 2003.
Reguant, J. 1997. Sria: Histria en imatges. 1894-1975. Angle Editorial: Manresa, Spain.
Reid, W. V. 2006. Nature: the many benefits of ecosystem services. Nature 443 (October): 749.
Reisner, M. 1993. Cadillac Desert: The American west and its disappearing water. Penguin Books: New York.
Repetto, R.C. (Ed). 2006. Punctuated Equilibrium and the Dynamics of U.S. Environmental Policy. Yale University
Press: New Haven, CT.
Royal Decree. 2003. Establishment of Spanish Drinking Water Quality Standards. 2003/140. 7 February 2003.
Ruggeri, J. 2008. Government investment in natural capital. Ecological Economics 68(6): 1723-1739.

200.
Sabater, S., E. Vilalta, A. Gaudes, H. Guasch, I. Muoz and A. Roman. 2003. Ecological implications of mass
growth of benthic cyanobacteria in rivers. Aquatic Microbial Ecology 32: 175-184.
Sachs, J.D. and W. V. Reid. 2006. Investments toward sustainable development. Science 312: 1002.
Sagoff, M. 2011. The quantification and valuation of ecosystem services. Ecological Economics 70: 497-502.
Sanders, R.A. 1984. Some determinants of urban forest structure. Urban Ecology 8: 13-27.
Sangenberg, J.H. and J. Settele 2010. Precisely incorrect? Monetising the value of ecosystem services. Ecological
Complexity 7: 327-337.
Santalla Torrens, E. 2008. Quan el vapor de la Burs Parl Angls: Impremta Pags: Barcelona.
Saur, D. 2003. Lights and shadows of urban water demand management: The case of the Metropolitan Region of
Barcelona. European Planning Studies 3(11): 229-243.
Schneider, D.W. 2011. Hybrid Nature: Sewage Treatment and the Contradictions of the Industrial Ecosystem. MIT
Press. Cambridge, MA.
Schneider, D.W. 2000. Local knowledge, environmental politics, and the founding of ecology in the United States:
Stephen Forbes and "the lake as a microcosm" (1887). Isis 4: 681-705.
Schneider, D.W. 1996. Enclosing the floodplain: Resource conflict on the Illinois River, 1880-1920. Environmental
History 1(2): 70-96.
Schrter, D. et al. 2005. Ecosystem Service Supply and Vulnerability to Global Change in Europe Science 310(25):
1333-1337.
Schultz, K. 2010. Being wrong: adventures in the margin of error. Harper Collins: New York, NY.
Scott, J.C. 1998. Seeing like a state: How certain schemes to improve the human condition have failed. Yale
University Press: New Haven, CT.
Seedang, S., A.G. Fernald, R.M. Adams and D.H. Landers. 2008. Economic analysis of water temperature reduction
practices in a large river floodplain: an exploratory study of the Willamette River, Oregon. River Research
and Applications 24: 941-959.
Service, R. F. 2006. Desalination freshens up. Science 313: 1088-1090.
Simpson, D.R. 2001. A Note of the Valuation of Ecosystem Services in Production. Resources for the Future
Discussion Paper. 01-16. Online: http://www.rff.org/documents/RFF-DP-01-16.pdf
Small, B. and N. Jollands. 2006. Technology and ecological economics: Promethean technology, Pandorian
potential. Ecological Economics 56: 343-358.
Sorlini, S. and C. Collivignarelli. 2005. Trihalomethane formation during chemical oxidation with chlorine, chlorine
dioxide and ozone of ten Italian natural waters. Desalination 176: 103-111.
Spirn, A.W. 1984. The Granite Garden: Urban Nature and Human Design. Basic Books: New York.
Srinivas M. B. 1999. What are trihalomethanes? On Tap (Spring): 18-19.
Stanton, T., Echavarria, M., Hamilton, K. and Ott, C. 2010. State of Watershed Payments: An Emerging
Marketplace. Ecosystem Marketplace. Available online:
http://www.foresttrends.org/documents/files/doc_2438.pdf
Stern, D.I. 1997. Limits to substitution and irreversibility in production and consumption: A neoclassical
interpretation of ecological economics Ecological Economics 21: 197-215.
Sweeney, B.W., T.L. Boff, J.K. Jackson, L.A. Kaplan, J.D. Newbold, C.J. Standley, W.C. Hession and P.J.
Horowitz. 2004. Riparian deforestation, stream narrowing and loss of ecosystem services. Proceedings of
the National Academy of Sciences 101(39): 14132-14137.
Swetnam, T.W., C.D. Allen and J.L. Betancourt. 1999. Applied Historical Ecology. Ecological Applications 9(4):
1189-1206.
Swyngedouw, E. 2005. Modernity and hybridity: Nature, Regnereracionismo and the production of the Spanish
waterscape 1890-1930. Annals of the Association of American Geographers 89(3): 443-465.
Tallis, H., R. Goldman, M. Uhl and B. Brosi. 2009. Integrating conservation and development in the field:
implementing ecosystem service projects. Frontiers in Ecology and the Environment 7(1):12-20.
doi:10.1890/080012.
Tello, E., and J. Ostos. 2011. Water consumption in Barcelona and its regional environmental imprint: a long-term
history (1717-2008). Regional Environmental Change 1-15.
Teuler, A. 2006. Eliminacin de contaminantes y otras sustancias presentes en aguas de consume mediante
tratamientos de membranas. XXVI Jornadas AEAS. La Corua, Spain. 8 June 2006.

201.
Theurer, F.D., I. Lines and T. Nelson. 1985. Interactions between riparian vegetation, water temperature and
salmonid habitat in the Tucannon River. Water Resources Bulletin 21: 53-64.
Theurer, F.D., K.A. Voos and W.J. Miller. 1984. Instream water temperature model. Instream flow information
Paper 16. US Fish and Wildlife Service FWS/OBS-85/15.vp.
Thorp, J. H., J.H. Flotmersch, M.D. Delong, A.F. Casper, M.C. Thoms, F. Ballantyne, B.S. Williams, B.J. ONeill
and S. Haase. 2010. Linking ecosystem services, rehabilitation, and river hydrogeomorphology. 2010.
Bioscience 60: 67-74.
Tonolla, D., V. Acua, U. Uehlinger, T. Frank and K. Tockner. 2010. Thermal Heterogeneity in River Floodplains.
Ecosystems. DOI 10.1007/s10021-010-9350-5.
Torgersen, C.E., R.N. Faux, B.A. McIntosh, N.J. Poage and D.J. Norton. 2001. Airborne thermal sensing for water
temperature assessment in rivers and streams. Remote Sensing of Environment 76: 386-398.
Toroz, I. and V. Uyak 2005. Seasonal variations of trihalomethanes (THMs) in water distribution networks of
Istanbul City. Desalination 176: 127-141.
Turner, R. K., J. Paavola, P. Cooper, S. Farber, V. Jessany and S. Georgion. 2003. Valuing Nature: lessons learned
and research direction. Ecological Economics 46: 493-510.
Tzoulas, K., K. Korpela, S. Venn, V. Yli-Pekonen, A. Kaierczak, J. Niemela and P. James 2007. Promoting
ecosystem and human health in urban areas using Green Infrastructure: A literature review. Landscape and
urban planning 8: 167-178.
USGS 2009. Surface Water and Water Quality Models Clearinghouse. United States Geological Service. Internet
[http://smig.usgs.gov/cgi-bin/SMIC/model_home_pages/model_home?selection=sntemp]. Accessed on 6
June 2009.
Valderi-Prez R., M. Lpez-Rodrguez and J.A. Ibez-Mengual. 2001. Characterizing an electrodialysis reversal
pilot plant. Desalination 137: 199-206.
Valero, F. and R. Arbs. 2010. Desalination of brackish river water using Electrodialysis Reversal (EDR) Control of
the THMs formation in the Barcelona (NE Spain) area. Desalination 253: 170-174.
Vzquez-Sue, E. E. Abarca, J. Carrera, B. Capino, D. Gmez, M. Pool, T. Sim, F. Batlle, J.M. Nierola and X.
Ibez. 2006. Groundwater modelling as a tool for the European Water Framework Directive (WFD)
application: The Llobregat case. Physics and Chemsitry of the Earth 31: 1015-1029.
Venkatachalam, L. 2004. The contingent valuation method: a review. Environmental Impact Assessment Review
24: 89-124.
Villanueva-Belmonte, C. 2003. Subproductes de la desinfecci de lAigua Potable i Cncer de Bufeta Urinria.
Doctoral Dissertation. Universitat Autonoma de Barcelona. Bellaterra, Spain.
Voltes i Bou, P. 1967. Historia del Abastecimiento de Aguas de Barcelona. Barcelona. Ed. Sociedad General de
Aguas de Barcelona: 198-199. Barcelona, Spain.
Walsh, C.J., A.H. Roy, J.W. Feminella, P.D. Cottingham, P.M. Groffman and R. P. Morgan 2005. The urban stream
syndrom: current knowledge and the search for a cure. Journal of the North American Benthological
Society 24(3): 706-723.
Walters, C. 1986. Adaptive Management of Renewable Resources. MacMillan Publishing: New York.
Watanabe, M., R.A. Adams, J. Wu, J.P. Bolte, M.M. Cox, S.L. Johnson, W.J. Liss, W.G. Boggess and J.L. Ebersole.
2005. Toward efficient riparian restoration: integrating economic, physical, and biological models. Journal
of Environmental Management 75: 93-104.
Webb B.W., D.M. Hannah, R.D. Moore, L.E. Brown, F. Nobilis. 2008. Recent advances in stream and river
temperature research. Hydrological Process 22: 90218.
White, R. 1995. The organic machine. Hill and Wang: New York.
Wilcox, B. P., M. K. Owens, W. A. Dugas, D.N. Ueckert,and C.R. Hart. 2006. Shrubs, streamflow, and the paradox
of scale. Hydrological Processes 20(15): 3245-59.
Worster, D. 1985. Rivers of Empire. Water, Aridity, and the Growth of the American West. Oxford University
Press: New York.
Wunder, S. 2005. Payments for environmental services: some nuts and bolds. Ocassional Paper No 42. Center for
International Forestry Research: Bogor, Indonesia.
Wunder, S. 2007. The efficiency of payments for environmental services in tropical conservation. Conservation
Biology 21: 48-58.

202.
Wunder, S., S. Engel and S. Pagiola. 2008. Taking stock: A comparative analysis of payments for environmental
services programs in developed and developing countries. Ecological Economics 65: 834-52.
Young, R.F. 2010. Managing municipal green space for ecosystem services. Urban forestry and urban greening 9:
313-321.
Zann, A. 2011. La ACA alcanza un preacuerdo para reestructurar 400 millones de su deuda. Expansin Catalua.
17 November 2011 5.
Zwieniecki, M.A. and M. Newton. 1999. Influence of streamside cover and stream features on temperature trends in
forested streams of Western Oregon. Western Journal of Applied Forestry 14: 106-113.

203.

Potrebbero piacerti anche